- Tidbits on Amazon's Web Services business
- One Way To Deal With Server Overheating
- Oracle Gets Big
- Who wins if Oracle gets BEA?
- Microsoft misses the SOA boat (again)
- WGA Server Outage Says Little About SaaS
- Server and Data Center Energy Efficiency
- Microsoft 'Shrink-Wrapped SOA' Coming in September
- NetworkWorld charts open source systems management
- When to Buy, When to Build
April 25, 2008 | Comments: (0)
Tidbits on Amazon's Web Services business
This Wired story has some very interesting comments on Amazon's Web Services (AWS) business. The first point was news to me:
"And the idea that AWS is mostly about wringing extra bucks (especially off-season) out of Amazon's data centers? "We've far exceeded the excess capacity of our internal system," Amazon's Jassy says. "That ship sailed 18 months ago."....
"I'd be surprised if no one else does this," Bezos says, pausing for effect. "It's a really good idea!" And there may be an ace up his sleeve. Any economist will tell you that a commodity business — storing and processing data, for instance — is a mug's game, with prices that plunge inevitably toward the cost of production (in the case of bits, pretty close to zero). That's music to Bezos' ears. "Commodity businesses don't scare us," he says. "We're experts at them. We've never had 35 or 40 percent margins like most tech companies.""
...
Wall Street's best guesses for AWS's 2007 revenue don't even reach $100 million.
I scoured Amazon's 2007 annual report to see if there was any additional data on the size of the AWS business. No luck. About all I could find is that Amazon's expenses for "Technology and content", which is where AWS expenses would be counted, were up to $715M ($818M if you include the cost of stock compensation to employees in the "technology & content" group). If the AWS business has grown beyond using the excess capacity of Amazon's internal systems, it wouldn't surprise me if 5-10% of that figure was associated with AWS. And based on the comment from Bezos, I'd probably uplift the expense estimate on AWS by < 25% to represent gross profit. All told, we'd be around [717M x ((5% + 10%)/2) x 125% =] $67M in revenue by my back of the envelope estimate. Not bad for 2007 (if my math is close to reality)...would love to know how big 2008 is considering that all the cool kids are using AWS.
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on April 25, 2008 02:48 PM
November 28, 2007 | Comments: (0)
One Way To Deal With Server Overheating
Looks like Microsoft has figured out one way to deal with data center overheating: Put the servers in Siberia.
Ok, I am sure there are other good reasons behind this possible move. But it does raise some interesting issues about thinking globally as you build out IT infrastructure. You need to think about where your web traffic is coming from, local skills, costs, access to low-cost power, reliability and a host of other issues. As we move to a more "on-demand" world for Enterprise 2.0 applications, these issues will become increasingly important.
Plus it also becomes a point of pride for local investments and doing business in other countries. Not to mention a possible training ground for developing internal IT talent.
Posted by Zack Urlocker on November 28, 2007 07:11 AM
November 23, 2007 | Comments: (0)
Last week's Oracle Open World event reminded me just how big Oracle has become. The conference pulled 43,000 people into Moscone Center and shut down traffic for several blocks around the convention center. While I would not say it was on scale with Comdex in its heyday, it was pretty impressive to see so many people attending a conference around just one company. Admittedly, this mega-event combines conferences from the many acquisitions Oracle has done in the last few years, such as PeopleSoft, Siebel and others.
There were some open source announcements at the conference, including Oracle's news of their Oracle VM offering, based on the Xen hypervisor also used by Sun, Citrix, Red Hat and pretty much everyone other than Microsoft and VMWare. CEO Larry Ellison was positive on the progress made by Oracle Unbreakable Linux, though with only 1500 customers, I'm not sure that it's making a dent in any existing Linux businesses. It's a good offering for current Oracle customers who want one-stop shopping, but I don't think too many companies beyond Oracle have done much in the way of supporting their variant of Red Hat. (And of course, that's a chicken and egg problem. Not enough users means no incentive for ISVs to test.)
Unfortunately, I missed the one open source panel that would have been interesting. Luckily, it's been blogged by Matt Anslett from the451 Group. (BTW, if you're not familiar with the451Group, I think they are the best analysts in the business. They get IT and they get open source. A rare combination!)
Scale matters in IT, especially at the infrastructure level. As Larry Ellison has said "they get big" over at Oracle. If you take away the marketing fluff, it's clear that Oracle understands scale. And being big doesn't necessarily mean boring. There are plenty of interesting technical issues involved in building large scale systems. Oracle is a partner with MySQL on the InnoDB storage engine and they've brought a lot of expertise to the InnoBase team to improve scale-out in the latest MySQL 5.1 release.
One nice side effect of a big conference is that they also get big-name talent at their customer appreciation party. As Lenny Kravitz said in his keynote, I mean, late night gig, "I don't know who's putting on this show, but y'all know how to throw a party." And I have the photos and videos to prove it.
Special thanks to Ken Jacobs (Dr. DBA) from Oracle for the hospitality!
Posted by Zack Urlocker on November 23, 2007 06:18 AM
October 12, 2007 | Comments: (0)
Man, what a wacky world software has become. Remember when $6.6 billion meant something? It's play money at this point.
Who wins and who loses if Oracle gets BEA?
Oracle: Probably
Oracle has never had a clear story about app servers and despite the successful "Fusion" branding it's never been totally clear on just what makes up Oracle's middleware offering. I think the big boost from buying BEA is that a large percentage of BEA users are also using Oracle. Oracle gets economy of scale out of their sales efforts and fills that piece of the puzzle.
BEA Customers: Maybe
Hard to tell. Weblogic is a good product (if very cumbersome) but I would be concerned that Oracle will just EOL the overlapping products (BPM for example) which BEA sold very hard to their existing user base. Hard to see this one having a happy ending but it seems like Oracle has mostly done right by their acquisition customers.
IBM: Yes
This is a big win for IBM, effectively leaving only small competitors for Websphere.
Red Hat/JBoss: Maybe
I am trying really hard to see this as a good thing for Red Hat. When you consider that the major JBoss server competitor will now be tied up for months as Oracle tries to absorb them it seems like the opportunity is there. On the other hand, it's starting to feel like Oracle is trying to kill Red Hat again. I like to think Red Hat can take advantage of this opportunity but it's a lot to deal with considering the JBoss machine hasn't been functioning on all cylinders. Add to that the fact that Oracle will now have a well-proven suite of SOA products and it gets a bit tougher.
Software Buyers: Nope
This is the biggest problem for me--the fact the choices in the marketplace are even slimmer than before with questionable benefits. It might be easier to buy from one vendor but that doesn't mean you'll like all the products.
Open Source: Yes
Anytime these behemoths get together there is huge risk for them screwing up the acquisition which means the door is open for open source.
Posted by Dave Rosenberg on October 12, 2007 01:24 PM
September 11, 2007 | Comments: (0)
Microsoft misses the SOA boat (again)
Microsoft just announced the release of BizTalk Server 2006 R2 (note to Microsoft, it's nearly 2008) which apparently is the cornerstone of their SOA strategy (please insert Nelson Muntz Ha-ha! related to MS and SOA.)
The availability of BizTalk Server 2006 R2 underscores Microsoft's long-term commitment to deliver extensive SOA and BPM capabilities as part of the overall Microsoft application platformFunny thing is I just did a Q&A all about how Microsoft has completely missed the point--and this proves that they continue to. They are just taking advantage of their captive market to convince them that SOA is all about Microsoft platforms and not about architectural advantage.
Posted by Dave Rosenberg on September 11, 2007 08:47 AM
August 29, 2007 | Comments: (0)
WGA Server Outage Says Little About SaaS
We've all read about the Windows Genuine Advantage server outage. Dvorak used the event to raise concerns about SaaS.
"All this proves is that these Web-based applications cannot be trusted."
I am by far the biggest supporter of SaaS, but I do see it as a market inevitability.
The WGA outage proves that when you don't design a system for five or six 9's uptime, you won't magically get that level of availability. As web apps move from "good enough" to "business critical", these types of outages will be outliers.
Here's another Dvorak conclusion:
"What is often lost in individual analyses of how to proceed with your data-processing needs is the concept of "being at the mercy of a single company." "
If you work for a medium to large enterprise, it's very likely that ADP, one of the oldest SaaS providers in the world, handles your payroll. Your company is very likely at the "mercy of a single company, ADP, when it comes to processing your bi-weekly salary and paying you on time". And yet businesses choose to use ADP.
Staying with the ADP example: I find it somewhat amusing that "fear of data loss or risk of your data being mistakenly viewed by a competitor in a multi-tenant SaaS environment" are inhibitors to broader SaaS adoption. Isn't your employee's salary & privacy concerns just as important as your customer lists or sales records? Sadly, maybe the answer is no.....
In any case, SaaS isn't going anywhere. It's been here for decades. A few SaaS-related outages simply make the case for more mainframes! ;-)
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on August 29, 2007 08:54 PM
August 07, 2007 | Comments: (0)
Server and Data Center Energy Efficiency
Slashdot reported that the US Environmental Protection Agency (EPA) delivered a study on Server and Data Center Energy Efficiency. Here is the 13 page summary and the full (133 pg) report. Here are a few highlights from the summary:
"The energy used by the nation’s servers and data centers is significant. It is estimated that this sector consumed about 61 billion kilowatt-hours (kWh) in 2006 (1.5 percent of total U.S. electricity consumption) for a total electricity cost of about $4.5 billion. This estimated level of electricity consumption is more than the electricity consumed by the nation’s color televisions and similar to the amount of electricity consumed by approximately 5.8 million average U.S. households (or about five percent of the total U.S. housing stock).""Under current efficiency trends, national energy consumption by servers and data centers could nearly double again in another five years (i.e., by 2011) to more than 100 billion kWh (Figure ES-1), representing a $7.4 billion annual electricity cost."
The EPA presents scenarios for energy use by servers & data centers in the US from 2007-2011. Based on increasing degrees of server consolidation, adoption of energy efficient servers and power management, the EPA predicts that the *annual* cost savings in 2005 dollars (i.e. excluding inflation) would be between $1.6 to $5.1 billion in the US. Yes, I know that $1.6B to $5.1B is a small figure when compared to the total US IT market spending on energy. I need to think about this more. I know that the "you're saving the planet" justification isn't going to fly with everyone (although it would with my sister-in-law :-). I believe that the benefits of cleaner IT will come down to cost savings even with higher rates of IT usage. I just don't have the data to back this statement yet.
Some of you may have caught the news last week that IBM consolidated over 3,900 servers onto about 30 mainframes running Linux. The move is expected to reduce server footprint by 85% and cut costs by $250M over 5 years. Very cool that Linux & mainframes are being used to save $$$ and reduce the environmental impact of IT.
This is another step in IBM's Project Big Green:
"The project, involving high-density computing systems that use server and storage virtualization, and energy-efficient power and cooling systems, is part of IBM's goal to double its data-center capacity by 2010 without increasing energy usage or carbon emissions."
It will be interesting to see how the server consolidation and virtualization trend impacts Linux adoption. On one hand, Linux on higher-end servers should be attractive to customers running Unix applications and seeking energy and cost savings. On the other hand, Linux has enjoyed its largest success on commodity servers that run at very low rates of utilization. Consolidating commodity servers onto a higher density, larger, more efficient server with higher utilization shouldn’t impact the number of Linux licenses. So maybe the net impact would be minimal on Linux adoption?
Thoughts?
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on August 7, 2007 09:01 PM
July 15, 2007 | Comments: (0)
Microsoft 'Shrink-Wrapped SOA' Coming in September
It's amazing to watch Microsoft continually be so clueless. 'Shrink-Wrapped SOA' is oxymoronic minus the oxy.
Regardless of the vendor, SOA "stacks" defeat the whole purpose of having a loosely-coupled architecture. Show me a stack vendor that allows for truly clean swapping of components with other vendors and I will eat my words. Of course even if a vendor's stack allows for substitute parts they are going to push toward theirs anyway.
Posted by Dave Rosenberg on July 15, 2007 07:08 PM
June 19, 2007 | Comments: (0)
NetworkWorld charts open source systems management
NetworkWorld compared the OSS system management products from Zenoss, Hyperic and Groundwork.
NWW results:
1. Zenoss-4.0
Pros: Excellent discovery of network resources; intuitive user interface.
Cons: Needs a visual design environment for customizing reports.
2. Hyperic-3.6
Pros: Impressive, complex thresholds; good reports.
Cons: Remediation is a very manual process.
3. Groundwork-3.1
Pros: Good, basic network monitoring.
Cons: Lacks dashboards; no ability to track Service Level Agreements.
One potentially interesting factoid is that each of these products has a different technology behind it. Zenoss is Python, Hyperic is Java and I think Groundwork is Perl with a bunch of PHP and Java. FiveRuns (not reviewed) was all written in Ruby so I guess you can solve the same problem in many different ways. (Apologies if I got those programming languages wrong.)
Posted by Dave Rosenberg on June 19, 2007 08:58 PM
June 17, 2007 | Comments: (0)
Dave Linthicum outlines six steps to follow to make a build vs. buy decision. (Spoiler: Ultimately you will probably need to do both.)
With service-oriented architecture (SOA), Ajax, the enterprise service bus (ESB) and other new technologies maturing, application architects and business analysts have tough decisions to make about whether to buy packaged applications, build them, or more likely, do some combination of the two.The buy-versus-build question is even more important now that we have infrastructure to support the mixing and matching of applications and services. Best practices are beginning to emerge, and sources for applications and services are rapidly multiplying, including open source, new SOA interfaces for traditional packaged enterprise applications and SaaS offerings. The SaaS vendors, in particular, are making new and innovative applications and services available at price points that were previously unheard of.
So, how do you know when to buy and when to build? There are no hard-and-fast rules, but this article presents a six-step approach to make sure you're making the right calls.
Posted by Dave Rosenberg on June 17, 2007 10:22 PM
June 12, 2007 | Comments: (0)
If LAMP is so easy, why isn't Java dead?
This is the question that Peter Yared asks in his blog, and it's fascinating given that he founded a company (ActiveGrid) based on the premise that Java was going to get shunted aside, as Stephen O'Grady notes. My company, Alfresco, is Java-based, so I'm biased in this, but it's interesting to hear the reasons Peter gives for Java's resilience:
1. Java has become easier.
2. It's really hard to move the world off a platform (Java) it is happy with.
On this latter point, Peter writes:
Even if it was quicker and cheaper to build a LAMP application than a Java application, the cost of getting a new software platform approved is pretty astronomical at a lot of enterprises. Adding staging servers, management software, security auditing, training staff in new technologies, etc. is very expensive. Linux and Apache have been enormously successful in the enterprise since they were brought in as a horizontal solution. Bringing in Linux to replace Solaris and running Apache for all static content significantly reduces costs. But running PHP or Python for some applications when an enterprise is already supporting Java and perhaps .Net is not viable for most enterprises, as the cost each added platform multiplies support costs.This has certainly been my experience. Enterprises are heavily invested in J2EE (or .Net), and just aren't throwing out that expertise. They're adding PHP expertise, but there is still the lion's share of the enterprise market for J2EE/.Net applications.
On the other side of the ledger, Google et al. are building massive businesses on the LAMP stack. Is LAMP the future of the enterprise and J2EE/.Net are its resilient past? I'm not sure, but based on the deals with which I'm involved, it's a past with a healthy future. I suspect there's ample room for both.
Posted by Matt Asay on June 12, 2007 10:27 AM
April 20, 2007 | Comments: (0)
Blackberry outage won't affect RIM monopoly
I admit that I have a Crackberry addiction. In fact I have to make sure that I don't look at it for hours on end lest I lose complete control. So when the network went down I was OK for a few hours, then my thumbs started to shake uncontrollably.
I don't really care that the service was down, I care that there was no information about how/why/when it would be fixed and that RIM would choose non-action as opposed to simple communications. Remember the JetBlue fiasco a few months ago? What if the RIM outage had been a major security breach? Would they have waited two days to tell the world that all their data had been compromised? Weak.
But the end result will be neglible (for now) as there is no real alternative to the Blackberry. This has always been Microsoft's approach to the desktop. Sooner or later real competition (don't tell me Treo) will appear and this kind of customer service will start to send people onto different devices.
Meanwhile RIM still offers the best devices and service. Where is the competition?
Posted by Dave Rosenberg on April 20, 2007 09:06 AM
April 12, 2007 | Comments: (0)
Google Apps Email-I am forced to give you the finger
We made a bold attempt to switch over to the Google Apps today and man what a world of pain we ran into with some of the stupidest "features" I have ever seen from a business application. The main issue is that you can't send POP mail to yourself from yourself and if you send from yourself to your alias it automatically archives. They claim this is to reduce inbox clutter. I claim it's moronic.
Sending from self to self = no pop
sending from self to alias = archive
The other amazing thing is that you can't reply-all and include yourself on the message thread. The header strips out your address and only includes the other recipients. I presume this is because Gmail is used to working within it's own system.
I can't believe that this is considered a good thing. We're going to give it a day to figure it out (changing MX records twice in one day does wonders for productivity FYI) but I can't see a way to make this work.
On the positive side I wanted to use Zimbra anyway, and I know it works just fine.
POP: I'm having problems downloading mail
Gmail Help Discussion
Posted by Dave Rosenberg on April 12, 2007 05:12 PM
April 05, 2007 | Comments: (0)
Switch to Google Apps (Verdict: Somewhat painful)
My infinite quest for operational efficiency continues to be elusive. Despite my love for Zimbra, our dev team really wanted to switch to Gmail and Google Apps for Enterprise. I do the Google Apps, but I am finding that if you have lots of mail already you are going to suffer the wrath of IMAP vs. Google.
So far it's been a bit painful-at least for those of us who have been using IMAP to store our mail. Right now, Google doesn't offer an import mechanism so you have to POP your mail into the inbox...which should be fine except Gmail can't seem to (not sure that it should) be able to find mail stored outside the inbox (which is the majority of my mail).
There are a few options for this situation:
-Try the Gmail loader script--this requires that you download all your mail and reforward it via Thunderbird or another Mbox like app
-Move mail into your inbox then POP it into Gmail
For now I have opted for the latter and I have to say it's an enormous pain in the ass. I am sure there is some other way to do this via 3rd party but in the meantime I truly hate this missing link. On the plus side, I feel like I could switch off Mac mail entirely, but still need offline capability.
Check out David Berlind's write-up.
Posted by Dave Rosenberg on April 5, 2007 08:19 PM
April 02, 2007 | Comments: (0)
Funding News: Vyatta raises $11m from Comcast
Maybe I shouldn't complain about my outrageous cable bill now that Comcast is investing in open source companies? Pretty interesting to see corporate behemoths investing in OSS technologies that help run their business.
Link:
Comcast money helps open source scare Cisco
Posted by Dave Rosenberg on April 2, 2007 08:08 AM
April 01, 2007 | Comments: (0)
Google Apps for Enterprise missing full-time SSL?
Having mostly recovered from our MuleCon event last week I looked at my to-do list and found that I had to be the one to setup the Google Apps Premier. After some minor difficulties (several "page not found" errors when adding accounts) I thought we were good to go.
I sent out email to the team to test drive the new Google mail and we immediately realized that the site shuffles you to a non-SSL page after login. I am sure this is for speed but it's a bit odd to think that corporate IT departments would suggest non-SSL browser based mail.
From our internal mailing list, Andrew pointed out:
Being a 2-year gmail user, I've been able to use the fully encrypted channel all the time via web: https://mail.google.com/mailThe difference from a standard URL is that once logged in with the above URL, one **stays** in SSL mode. The default URL drops out of SSL and goes into a plain unencrypted HTTP mode.
I am greatly surprised to have not found the same behavior for our new MuleSource account. None of my URL variations worked. Google groups research only revealed similar questions without any answer.
It's probably good to give their support a call (it's free with Premier accounts) and find out what's going on. Otherwise free gmail users are in a better position then paying customers (?!).
For the moment I am going to chalk this up to a configuration issue, but it's a big one that will prevent large customers from using these services. This is exactly why people are afraid of hosted apps.
Posted by Dave Rosenberg on April 1, 2007 04:52 PM
March 06, 2007 | Comments: (0)
Open Source and Systems Management: Perfect Together
In conjunction with my post about the incredibly foolish argument against open source by Roger Greene of Ipswitch, I invited Javier Soltero, CEO of Hyperic to comment.
--------
Open source is not only a reality, it's a better option for both vendor and customer. Customers want to trust their software decisions, and that is hard to do when discovery and due diligence is clouded by the smoke and mirrors of smarmy sales people and pretty powerpoint. Customers want to kick the tires, they want to take it for a test drive, and they want to do it themselves. Old school closed source vendors can't stand this process, as it drags out the sales cycle, increases the cost of sale, and means they are going to have to work for that sale. Good open source products with strong are designed for the customers to realize value on their own, and support each other in the process - including the case for the vendor to do product development that echos what the market wants. This means open source vendors realize dramatic cost savings for support, sales and more effective product development. These cost savings are passed directly on to the customer.
Also, because products are designed to be installed and operated by the people that need them - implementation costs plummet. Let me illuminate you with two examples. A favorite story around the Hyperic office is that two years ago, one of our employees was doing product management for an old school, closed source vendor. As good product managers do, they go out and talk and watch their customers build and use their product implementations. In this case it was a large telco in Europe. In this process, she witnessed an entire floor of 120 Accenture consultants working on a project. It wasn't directly working on her software, it was working on HP OpenView - to manage it. 120 Accenture consultants had been there 6 months and they still weren't done! I don't know what the software cost was, but the old rule of 3x for implementation services had to be blown away. Fast forward to last week, and we had a customer who had decided based on community reputation to try our product. They were looking at doing a system replacement with our new 3.0 product. The administrator asked if he could come into our office ask specific questions about phasing the old product out. Because he was around the corner, we did this in person. After 1 hour of a barrage of questions, his response was, "I definitely made this out to be more complex then actually is. I guess I've just been used to this stuff being hard to implement. This is going to be cool."
So, let's go back to Roger's logic about being incented to cost more in service. While I didn't see the bill myself, with the reputation of HP and 120 tier one consultants for 6+ months, I think its safe to say the cost was 7 figures, and probably 8. If Hyperic did anything like that, we'd be out of business. We'd need a legion of sales and marketing pros to shove the cost down customer's throats. Instead, we have a relatively small team of sales and marketing (think single digits), and customers flocking to us. They've implemented pilots themselves, without our involvement or any expensive consultants, and now want to pay for indemnity and support. And by pay, they are only paying a few hundred dollars per server. Nothing like that poor telco.
As for open source adoption, the adoption numbers for USA and Europe are around 80%. That means 80% of all businesses have open source somewhere in their infrastructure. This article should illustrate that point nicely. http://weblog.infoworld.com/openresource/archives/2007/01/open_source_usa.html. I recently heard that the European Union has a policy that all open source options have to be exhausted before they will even look at proprietary. For anyone still comfy in their old world software bubble, I suggest you wake up. Most vendors with a clue are going open source. Just check out this research from the 451 Group: http://the451.com/caos/caos_detail.php?icid=312.
And don't even THINK about that other lazy mindset that open source is 80% of the functionality for 20% of the cost. I can't think of better proof that Open Source is the hotbed of innovation then to congratulate my peers at OpenNMS for kicking some old school, arrogant, sluggish and ineffective, proprietary software a$$. Kudos to you my friends for having the passion and innovation to push the market farther and better serve your customers and community.
Posted by Dave Rosenberg on March 6, 2007 12:52 PM
March 05, 2007 | Comments: (0)
OpenNMS bests OpenView and Tivoli while Ipswitch spreads the FUD
Chalk up another victory for OSS over proprietary. OpenNMS beat out both OpenView and Tivoli in the SearchNetworking Product Leadership Awards. I wonder if that will shut up this ridiculous FUD from Ipswitch "Don't trust your network to open source."
I let Travis take the shots at this foolishness...wake up, Ipswitch, you are late to the FUD train. Javier...anything from you?
Myth #1 - Open Source is free - According to Greene, downloading open source from the Internet and then customizing to your environment "often is not a good use of your time." Greene adds that he'd "rather pay an upfront fee for software that does what I need and doesn't have any high-cost labor attached to it."Hmmm ... what about the fact that proprietary software (and *especially* network monitoring and management products) are often tremendously difficult to install / configure / maintain ongoing? How is being held hostage to a vendor for support / installation / configuration preferable? And how is being tied to a predetermined feature set preferable to having the ability to customize an open source approach solution to meet your environment's needs?
Myth #2 - Bug fixes are faster and less expensive in an open source environment - the second "myth" that Greene exposes around open source is the notion that there are thousands of developers sitting at home contributing labor for free. Greene suggests that most of the contributing vendors are typically employed by large vendors … and that "even when those individuals generously offer their time for free, can you really afford to wait for one to agree with you on the urgency of action if your network is down."Hmmm ...so it's better NOT to have access to the source code when you have a bug? It's preferable to have to open a help ticket with the vendor and wait in line? It's better NOT to have general visibility into the bugs and issues being reported by the members of the user community?
Myth #3 - Your IT staff can buy a 'raw' tool and shape it to their needs - Greene's last point is that the industry has moved away from the "classic open source" model where folks download raw open source and customize to their needs - and to more of a commercial open source model, where organizations are leveraging open source distribution as a way to sell services.
Posted by Dave Rosenberg on March 5, 2007 03:04 PM
March 01, 2007 | Comments: (0)
Why Software as a Service needs Open Source
I have an article running on TheServerSide.com about the challenges of integrating your enterprise infrastructure with SaaS apps. I think that the main point is that if not properly thought through, your SaaS apps just become silos.
Salesforce.com saw this very early and built some fantastic APIs, but typically those APIs are used to map fields rather than do full scale integration. This was one of the main things we heard from several very-large Mule users--that they wanted the ability to connect multiple apps, databases and other online content and drive it in and out of Salesforce.com.
So why does open source make sense? Many SaaS apps are built on OSS components, and if you are building a SaaS app it's clearly helpful when you can tweak code to be able to support multi-tenant architectures. When we are talking about integration of SaaS and internal apps, the ability to modify a piece of software to meet your exact needs is a huge help. It's what's driving open source into the heart of the SOA.
Posted by Dave Rosenberg on March 1, 2007 03:49 PM
February 28, 2007 | Comments: (0)
Zmanda throws down with Veritas
I spoke with Zmanda CEO Chander Kant last week about the company's growth and some new things they have been working on. I first met Chander at the MySQL conference about a year ago. Since then the company has expanded its support for MySQL and is now starting to target other applications like Oracle and SAP.
Today Zmanda announced a program that I think is an interesting marketing move.
Beginning March 1st and through April 15th of 2007, the leader in open source backup Zmanda will offer existing Veritas NetBackup customers the ability to switch to its brand new Amanda Enterprise Edition 2.6 with Zmanda Management Console for free!Simply show your Zmanda Account Manager the current renewal quote for Veritas NetBackup and Zmanda will set you up with a year long subscription including software and support for Amanda Enterprise 2.6. Yes, its that simple - just like installing and using our products.
There are two aspects of this that I like:
1. By calling out Veritas they make it clear who they compete with, rather than just saying backup
2. Since the support infrastructure is already in place the incremental cost associated with doing this is very low. And once you start using the management console it's unlikely that you would swap it out for an expensive solution from Veritas.
Chander tells me the management console is very different from the other backup vendors--typically these consoles are Java-based and very heavy. The Zmanda console is browser-based and takes advantage of other web properties. They will also soon publish an API for Amanda for 3rd party developers to do direct integration.
Previously:
Zmanda-Open Source Data Protection
Posted by Dave Rosenberg on February 28, 2007 08:16 PM
February 26, 2007 | Comments: (0)
Open X-change lands super-size deal
Businessweek is reporting that 1&1 Internet, the world's largest Web hosting company, said it will roll out 1 million e-mail accounts this year running on Open-Xchange's open-source software.
That's a lot of mailboxes, but I wonder if it's actually the largest deployment of open source software.
Posted by Dave Rosenberg on February 26, 2007 08:11 AM
February 23, 2007 | Comments: (0)
SAP to offer highly uncustomizable apps for mid-market
SAP continues to avoid the learnings of open source...How can customers not flock to a product that is billed as "highly standardized" with "less choice" of options?
Kagermann referred to A1S as an "entirely new product" that will not be a hosted version of the company's All-in-One product. All-in-One is a slimmed down version of mySAP Business Suite, which is designed for large enterprises.The hosted offering will be a highly preconfigured, "off-the-shelf" product that gives companies a range of applications with limited flexibility to tweak their functionality, he said.
"To drive the price down, the product must be highly standardized," Kagermann said. The need to offer a set of standardized functions, he said, also means "less choice" of options to customize the product, he added.
By narrowly limiting product customization and thus complexity, SAP and its partners will be able to deploy the offering much faster, Kagermann said. Fewer consultants will be necessary, resulting in lower costs.
Previously:
SAP: We spend lots of money, therefore we provide value
Posted by Dave Rosenberg on February 23, 2007 11:14 AM
February 08, 2007 | Comments: (0)
Q&A with Bill Cook, CEO of Greenplum
This week open source data warehousing provider Greenplum has hit its stride, announcing $19M in financing and adding Sierra Ventures to it's list of investors, hiring a new CEO from Sun to help it maximize the partnership the company's announced last year, and signing its first data warehouse appliance customer with Sun, Philippines-based wireless service provider Smart Communications.
Recently, I had that chance to chat with Bill Cook, Greenplum's incoming CEO about what attracted him to the opportunity, and how he thinks the company's open model will affect the high-end data warehousing market.
What made you decide to join the Greenplum team?
When I started talking with Greenplum, it really reminded me of Sun in the early days-a large and growing market opportunity, disruptive technology and world-class talent. I was especially attracted to Greenplum's software-only focus, open-source community-centric approach, and partnership model.
How will Greenplum-and Sun-be affected by recent moves into this space by HP and others?
For a long time the DW market has been in need of a shake up, and we're excited at some of the changes taking place. We think that more choice is ultimately going to benefit customers, and we think that the Greenplum offering is uniquely able to solve the problems that more traditional approaches have been unable to address.
What makes you think a newcomer like Greenplum can upset the high-end data warehousing market?
Well, for one, we can do it better, faster, and for less money than the alternatives. Period. Greenplum has attracted an amazing team of world-class sales, marketing and engineering people who really know what they're doing. The combination of the power of open source with channels of tier-one IT vendors like Sun to deliver best-in-class turnkey appliances is the best approach to driving adoption.
Greenplum's software-only business model is unique to this market. What makes you believe it is the right approach?
Our feedback from customers is that our software-only model is the right one. Our focus is to be the best software company we can be. We believe that by leveraging the systems innovation, global infrastructure, and field operations of the major systems vendors like Sun we can achieve faster adoption than the niche, stand alone data warehouse appliance vendors can keep up with.
Do you have plans to partner with other large systems companies?
The Sun partnership is very important to us, and we are focusing our efforts on maximizing the relationship. Over time, we hope to work with other platform vendors as well.
What can we expect to see from Greenplum this year?
We're seeing the demand globally. Sun / Greenplum are working on opportunities with the largest retailers, banks, telecom companies and Internet service providers services around the world. You can expect to see additional customer announcements from Sun and Greenplum in the coming months.
Posted by Dave Rosenberg on February 8, 2007 10:33 AM
February 07, 2007 | Comments: (0)
Credit Suisse Conference: Benioff is Bullish (Surprised?)
I'm at the Credit Suisse Disruptive Technology Conference today sitting in the Salesforce.com keynote. Not surprisingly they are pretty bullish about growth of on-demand.
On-demand growth stats:
IDC: 31% growth
McKinsey: 61% growth
According to Mark Benioff, everything is going on-demand and multi-tenancy is the core to the shift. The multi-tenant architecture is fundamentally different and you have to build it from the ground-up. This is obviously not for everyone, but this approach allows for application scale that is not achievable otherwise. Salesforce is currently delivering 60 million daily transactions spiking to 70 million recently.
Revenue is rising consistently as is customer growth, with over 550,000 paying subscribers (October 2006.) They are targeting all market segments, SMB etc. and going for big guns like Cisco who have about 15,000 seats. The latest mega-deal is Dell for 15,000 seats. Neither Oracle or SAP has announced that size deal in at least 2 years.
The Salesforce focus for the immediate future is extending the platform via Apex and AppExchange.
Posted by Dave Rosenberg on February 7, 2007 09:45 AM
January 30, 2007 | Comments: (0)
Via Matthew Aslett at CBR:
"We are planning to go public," Mickos told Computer Business Review in an exclusive interview, adding that the Uppsala, Sweden-based database management vendor is in no hurry to go public after raising $18.5m in Series C funding this time last year and $39m in total.
When it does go public, MySQL will be one of only a handful of open source vendors to do so. Red Hat, VA Linux (now VA Software), and Caldera (now SCO Group) went out in the boom. RHAT is the only stock doing well at this point.
Posted by Dave Rosenberg on January 30, 2007 12:22 PM
January 29, 2007 | Comments: (0)
Oracle may launch support for MySQL
Matthew Aslett spoke to MySQL CEO Marten Mickos this morning:
Here are just two things of note: Oracle has suggested it will offer support for the MySQL code, undercutting the company, and Oracle is already distributing the open source database management system.
Note that Oracle is already distributing MySQL in their Linux offering.
Posted by Dave Rosenberg on January 29, 2007 12:27 PM
January 26, 2007 | Comments: (0)
We finished up our meetings in Japan with an IT Architect Magazine interview and our presentation on Mule and open source for SOA.
Consumer technology here is in a constant state of flux and innovation while enterprise technologies are quite "conservative." Which in layman's terms means resistant to change. Our hosts from OGIS described an undercurrent of interest in both SOA and open source, but felt that the big vendors may have pushed the SOA market in the wrong direction with the introduction of things like BPM tools which looked nice but were largely non-functional in relation to the infrastructure in place already. This is not unlike what's happened in the rest of the world where the hype of SOA is only now starting to be realized...fortunately for me it's being realized on open source components like Mule.
OGIS built a baseline for full-scale SOA with a mix of open source and proprietary tools (see diagram) which seems to be the right approach—start incrementally at the core and bring in the other components on top of the base (in this case the ESB.) What's interesting is that everyone has seen the ESB diagram that shows one giant bus with lots of things connected but what we're seeing (and OGIS found) is that you need many smaller ESBs to create a fabric. One very cool thing they did (and why open source makes sense for SOA) is they wrote a module to connect their homegrown Excel/VBA application to Mule via SOAP. None of us had seen that one before. Very cool stuff.
Side note: Somehow I think I have gained weight here, adding to my already overwhelming feeling of gigantism. Must be the fact that I have eaten ramen or udon twice a day since Monday.
Posted by Dave Rosenberg on January 26, 2007 01:29 AM
January 22, 2007 | Comments: (0)
On Open Source IT Systems Management - Q&A with Nora Denzel (formerly of HP OpenView fame)
I recently had a chance to chat with Nora Denzel, former Senior Vice President and General Manager of the Software Global Business Unit of Hewlett Packard, to talk shop about IT operations management and open source's role / opportunity in the space. Here's a snippet of the QA.
Open Sources: How big is the IT monitoring/management space today? Who are the major players that occupy it?
Denzel: In 2006, Gartner released a study that pegged the IT operations management software market at about $7 billion a year with more than 50% of the number going towards IT infrastructure monitoring and management. The dominant legacy players that offer proprietary monitoring solutions are the usual suspects - HP OpenView, IBM Tivoli, CA Unicenter, and to a lesser extent BMC Patrol. Their software is filled to the gills with features, which often goes unutilized I'd like to point out, and is very expensive.
Open Sources: What role do you see open source solutions playing in the IT infrastructure space that has previously been dominated by the "Big 4"?
Denzel: Obviously, there's a significant cost savings with open source solutions compared to proprietary tools. And as most mid-market organizations (including educational institutions) can't afford to shop at the "Big 4" store, it opens up a big market opportunity for open source. Add to that the flexibility and extensibility of open source products that allow companies to "right-size" and customize their IT monitoring and management solutions to fit their specific needs. Being able to do this at a low cost had been a pipe dream for SMB's and smaller enterprises, since most of the solutions out there cater to the largest enterprises with the deepest pockets. The availability of open source options is changing that.
Open Sources: How do you see open source software disrupting the IT monitoring/management space?
Denzel: As customers embrace more and more low cost enterprise-grade open source IT infrastructure and network monitoring solutions, the "assess" layer of the IT operations management stack has essentially become commoditized. By "assess", I am referring to the collection and correlation (or monitoring) of raw data so that it can be of use to system administrators to make decisions. There is no longer differentiation with the monitoring, or "assess", level of management. Inflexible, expensive proprietary solutions from the "Big 4" vendors simply do not provide enough additional value, in comparison to open source alternatives, to justify the magnitude more cost. In fact, studies show that customers who use the expensive proprietary solutions use less than half of the product's features even though they're paying for the feature bloat.
This "commodity layer" line of thinking has historical precedent - Apache with web servers, JBoss with application servers, SugarCRM with CRM, MySQL for the RDBMS industry, and Linux in the datacenter. It also makes sense given the current trend in organizations to drive down infrastructure costs while focusing resources on initiatives that ultimately grow the bottom line.
Open Sources: What do you think about the future of open source in this space? What do you see happening over time?
Denzel: Already, research and development / marketing dollars invested by the "Big 4" vendors are not being devoted to enhancing the infrastructure monitoring software within their IT operations management suite. Rather, the dollars flow up the stack in the areas of IT service and business processes - areas where open source solutions are nascent and far from mature. It is at these "advise" and "act" layers where large proprietary vendors add real value for customers. Their R&D investment reflects that.
I predict a "draining the cost pool" for customers and a paradigm shift in the way they view management solutions. Big fish vendors will be pushed down to the deep end where they can add value while the adoption of best of breed open source solutions will lower the customer cost water line in the shallow end. Inevitably, as customers realize the commoditization effect of IT infrastructure monitoring due to the economic disruption of open source alternatives, the large proprietary vendors will follow suit by focusing on the upper layers of IT operations management, where prices are protected and open source alternatives are not ready for prime time. Open source IT infrastructure monitoring solutions can then serve as feeders to the upstream proprietary solutions.
Open Sources: What is the biggest challenges facing open source IT infrastructure solutions at present?
Denzel: I think they face many of the same challenges all open source tools and companies face -- the perception of limited / poor functionality as well as inferior documentation and support services that you'd expect from a commercial vendor.
I also think the "sex appeal" or buzz associated with open source IT infrastructure monitoring solutions still pales in comparison to other areas where open source technologies have forged further into the mainstream - the "P" languages, application and web servers, databases, Linux commodity boxes, etc. The technology may be fantastic and scalable and a legitimate alternative to proprietary stalwarts, but there's always the tendency of people to follow the herd and go with what they know - it's the whole tipping-point idea - and I think it's not too far off.
Posted by Dave Rosenberg on January 22, 2007 03:50 PM
January 04, 2007 | Comments: (0)
Rather than using email aliases that don't keep an archive of developer mailing list items, we made the switch this week to Google Groups, which it now appears we have broken. One of our guys (a group "owner") is now banned from the group for having used the service too much.
There seems to be a limit to the number of posts you can do in a hour (though there is zero information about that number) which basically leaves you dead in the water. I am guessing its some kind of spam prevention, but this a private group with only 15 members.
UPDATED: Seems that the offender did a dump of his old emails which might explain it--though it would be nice to have any idea what happened. And yes I checked the Groups group.
This of course is the danger of things that are free--despite the fact that we really like the features, the lack of information (especially when things go wrong) make it impossible to rely on for real-world usage.
Any suggestions on services that do private mailing lists (we really don't want to run our own) would be welcome--as would feedback from Google as to what the difference might be if the product were not in perpetual beta.
Posted by Dave Rosenberg on January 4, 2007 09:21 PM
November 18, 2006 | Comments: (0)
Where is the free wi-fi in Paris?
After coming in from NYC and London, where free wi-fi is near ubiquitous, I am paying 30 euro for a day of internet access in Paris. Ridiculous.
Anyone know anything about the broadband here? I also noticed far fewer mobile phone stores than in London.
Posted by Dave Rosenberg on November 18, 2006 05:07 AM
November 15, 2006 | Comments: (0)
The evolution of email...begins with its past
The best way to usher in a new world is often to coexist well with the old world.
This is one reason that Centeris is an interesting company, allowing enterprises to move to Linux without giving up the handholding of Microsoft's server management interface.
It's also one reason I like PostPath so much. PostPath promises a revolution in corporate messaging, but it assumes the world as it is (140 million Exchange mailboxes), not the ideal world (Linux messaging everywhere). PostPath offers true drop-in compatibility with Exchange, such that an enterprise can scale out its Exchange infrastructure with...PostPath (at a fraction of the cost (both acquisition and TCO), up to 20X better performance, and significantly easier administration).
What's not to love?
PostPath recently got even better, adding its Remote Office Edition. I stopped by PostPath for a demo today, and was impressed by what I saw. Here's what it's all about (in Duncan's words):
Historically, customers with remote offices have found themselves on the horns of a dilemma with Exchange. Deploy remotely (distributed), and you drive administration and management costs, as well tending to end up with an unreliable service since it is hard to maintain Exchange consistently and effectively in remote offices. Deploy at a NOC (centrally) [Network Operation Center], and you drive up bandwidth costs and reduce the quality of the user experience with high latency.In short, PostPath offers real high availability that is functionally better than what Microsoft Exchange offers, and at a dramatically lower cost. Very cool, Duncan. Now I just want it under a different license. ;-)PostPath Remote Office Edition solves this problem:
- Enables deployment of low-maintenance server at a remote office
- High performance enables low-cost hardware + storage
- Replicates the remote server back to the NOC using standard Linux - DRBD
- IT manages the replicate copy at the NOC - even does backup there
- Can turn on high-availability (HA) at the NOC if desired
- With HA, if remote office hardware fails, failover takes less than one minute
- WAN bandwidth costs minimized by local service for remote users
- Remote users get a high performance, high reliability solution (i.e., they're happy)
- Administration is simple and centralized (i.e., IT is happy)
Posted by Matt Asay on November 15, 2006 09:39 PM
November 10, 2006 | Comments: (0)
VMware Ace 2.0-an answer to network guests?
VMware has what could prove to be a great answer to concerns about guest access to corporate networks. Ace 2.0 will allow users to connect to networks via a VM image that is managed by the network admin. Overall a much smarter design then Microsoft InfoCard.
The product, which was demonstrated in alpha at VMworld this week, allows administrators to distribute pre-packaged virtual machines to users in a secure manner. This means, for instance, that contractors can be allowed to attach to the enterprise network using their own laptops but only via the ACE VM.
Link: VMware reveals new product features
Posted by Dave Rosenberg on November 10, 2006 03:43 PM
October 30, 2006 | Comments: (0)
Moving from open source "project" to "company"
SandHill.com is running an opinion piece I submitted "Launching an Open Source Business" which outlines some of what we went through in our funding process.
It's a decent read despite the fact that I wrote it after a severe bout of food poisoning last week. It includes a short list of things to consider during the funding process. The most important thing is that you must be the expert on your product and your market.
Posted by Dave Rosenberg on October 30, 2006 07:44 PM
October 30, 2006 | Comments: (0)
Open source keeps your options open
Mule project founder Ross Mason has an article running over on Dr. Dobbs called "Five Ways Vendors Gouge Customers on Integration."
1. Up-Front Coding, Proprietary Toolkits
2. Astronomical Licensing Costs
3. Ongoing Services/Support Costs
4. Getting Locked Into a Prescribed Approach
5. Feature Overshot
And while integration is clearly an expensive pain, the five points he elucidates are applicable to pretty much any market segment dominated by proprietary vendors.
Posted by Dave Rosenberg on October 30, 2006 07:38 PM
September 11, 2006 | Comments: (0)
SpikeSource raising $21m B round?
The Alarm:Clock says that SpikeSource is raising a $21m B round but the link seems to be broken right now. I can't confirm or deny and have no other information. I posted a screenshot so that no one freaks and blames me for anything.
SpikeSource raised $12m just over a year ago. If I may quote Homer Simpson "sweet merciful crap!" That is a lot of dough to burn through in one year and a few months. I am sure they got a whopping valuation despite little in the way of revenue. It's a great time to be an open source company :>
Posted by Dave Rosenberg on September 11, 2006 09:49 PM
September 11, 2006 | Comments: (0)
Anyone know the URL of SAP's New Expert Blog?
I noticed this story about SAP's New Expert Blog which I was interested in visting based on the article. And yet there is no mention of the URL, the name of the site or the authors making it as mysterious as the druids.
Yolton said that the site, quietly introduced fewer than four months ago, already has 30,000 members providing best practices and expertise on designing, developing, deploying and optimizing business processes.
Any help would be appreciated. I am finding myself more intrigued by SAP all the time and any chance to understand the company is welcome. It's a bizarre choice (mistake?) that the author wouldn't publish the URL and the interviewee wouldn't mention the name of the site. FYI-it might be this blog.
Posted by Dave Rosenberg on September 11, 2006 09:10 PM
August 01, 2006 | Comments: (0)
XenSource gets another black-eye
Yesterday Red Hat said they were having issues with XenSource and now Oracle concurs.
Two quotes from Oracle VP Bob Shimp:
"We certainly believe in one simple universal way to integrate a variety of virtualization solutions, and that is the way that Andrew Morton [the maintainer of the stable Linux kernel] wants to go""I can say that Oracle is losing its patience over this issue and we are going to be pushing harder and harder on everybody to come to the table with a realistic solution,"
Andrew Morton has noted on a number of occasions that he has been waiting for XenSource to deliver development to the kernel. Will XenSource blow their opportunity?
Posted by Dave Rosenberg on August 1, 2006 09:09 AM
July 31, 2006 | Comments: (0)
Why Red Hat says Xen isn't ready yet
News.com reported earlier today on Red Hat VP Alex Pinchev's comments on XenSource and why Red Hat isn't rolling it into RHEL just yet.
"XenSource is not stable yet, it's not ready for the enterprise," Pinchev told CNET News.com sister site ZDNet Australia on Monday.
"We don't feel that XenSource is stable enough to address banking, telco or any other enterprise customer, so until we are comfortable, we will not release it," he added.
Now The Register adds detail from an Illuminata report:
1. The new "credit scheduler" was just checked into the xen-unstable tree earlier this month. SLE10 shipped with two other schedulers (BVT and sEDF). According to this posting by Keir Fraser (one of the original developers of Xen): "With the new credit scheduler checked into the xen-unstable tree, I wouldn't recommend to use either BVT or SEDF. They're both buggy. The new scheduler is supported by us, automatically load balances on SMP systems, and has a simpler administrator interface. Once credit scheduler has demonstrated its stability we'll most likely remove the other two schedulers."
2. The SLE10 installer is primarily just for SUSE guests at this point. "...for Windows you have to use the manual approach by editing config files and changing entries to boot from CD/HD."
3. Performance profiling (xenoprof) is still in flux. e.g., it doesn't save or restore performance counter state on context switching (so it's difficult to profile performance within guest OSs).
4. According to Red Hat: "Xen security is an issue [that] Xen currently has no concept of: opening up the migration port, for example, will expose all of the guests to anybody on the network. (Imagine driving a server truck into a corporate parking lot, migrating all the instances to your servers, and driving off ;-).)"
This whole thing is a bit confusing, especially as I attended the Red Hat Integrated Virtualization launch where XenSource and Red Hat were all chummy. Don't ask me. I just blog here.
Posted by Dave Rosenberg on July 31, 2006 07:49 PM
July 05, 2006 | Comments: (0)
Birthplace of the cellphone to be razed
Off-topic: Back in 1996 I started my first "real" job in the hallowed halls of Bell Labs in Murray Hill. It was a total old-school lab environment to the point where we had bunsen burner gas valve things sticking out of the walls in pretty much every office. I used to see Dennis Ritchie in the hallway. It was a trip. In late 1997 I switched to a different business unit and worked out of the Holmdel building that is now being razed by a developer. All of our offices in that building (except the supreme executive wing) I think were used for operators at one point. Total isolation, windows only at the ends of the hallway, but a cool design with a giant atrium.
Designed and erected between 1957 and 1962 by the inimitable and infamous Eero Saarinen, Holmdel is former home to Bell Labs' optical transmission, microwave, and wireless work, including the High-Speed Networks Research Department, High Speed Mobile Data Research Department, and Data Networking Systems Research Department. It was Holmdel's Wireless Research Laboratory, however, and the work Richard Frenkiel and Joel Engel that ranks among all Bell Labs' most notable contributions. In the early sixties Frenkeil and Engeld led a team of over 200 engineers to develop the first cellular wireless voice transmission technology, and eventually created AMPS (Advanced Mobile Phone System), the first and one of the most widely deployed cellphone technologies (still active even today in many parts of rural America). Holmdel is effectively the birthplace of global wireless movement, possibly the most crucial communications development of the 20th century, the internet notwithstanding.
Sure all the tech stuff is cool, but the real interesting story is about a couple of scientists who hurled themselves off the top floor when some lifelong experiment went awry (maybe someone remembers?) Anyway, it's a damn shame to see what happened to Bell Labs, and a real waste of a legendary workspace. The Engadget guys are taking up the cause to save the building. I support them wholeheartedly.
Posted by Dave Rosenberg on July 5, 2006 07:17 PM
July 02, 2006 | Comments: (0)
DSL down-being offline creates productivity death spiral
Long ago I accepted that as a technology early-adopter I would suffer more then most normal people. This week has resulted in a perfect storm of my DSL going down (or at least the modem dying) and in an attempt at using Comcast internet, finding that the modem they shipped is defective (says their call center), so after 1 hour of usage, that too is now dead. Two broadband connections and no internet. I am stealing my neighbors' wifi to write this post. Pathetic.
Even more pathetic is the fact that I am now entrenched in an online tool called Basecamp that has no offline component and therefore I can't access or update anything I need to. Basecamp itself has some other issues; we are having a really hard time dealing with the fact that there is no sense of workflow built-in and the complete lack of seach functionality. We're going to move JIRA as soon as we have a server available. I still like Basecamp but it doesn't match the needs of every company.
Back the connectivity issue-I wrote about Zimbra recently and their realization that they need to have an offline version of their application. The same goes for all the Web 2.0 apps. I would say that I live online as much as or more than most. When I am disconnected my reliance on these web-only services affects my already meager productivity. And while I recognize the difficulties in developing software for multiple desktop platforms there needs to a balance between the online and offline worlds. This offline component is still important no matter how much bandwidth we have.
Link:
LinkedIn down: Proof that 'the offline problem' needs solving
Posted by Dave Rosenberg on July 2, 2006 03:01 PM
June 26, 2006 | Comments: (0)
As I browsed the news this morning I noticed that ActiveGrid's new release will run on both LAMP and J2EE deployments. I asked CEO Peter Yared to give us an update.
A few months ago, I posted an open letter to Jonathan Schwartz, then President and COO of Sun, asking him to open source Java so that the JVM would attract the PHP, Python, and Perl open source communities and bring the Java and LAMP worlds together with a single VM.Sun announced at JavaOne last month that they would open source Java. While no one is quite sure what the license will be, we at ActiveGrid have taken this to heart and have started contributing to the Jython project, mainly by updating Jython to Python 2.3 (from Python 2.1, where it had been languishing ever since Microsoft hired Jim Hugunin to make Python run better on the .net CLR).
With ActiveGrid Studio v2 and ActiveGrid Server v2, our LAMP-based technology will run within a Java server. All of our customers have found it incredibly easy to build new, Web 2.0 style applications that integrate with their enterprise backends, but have found it challening to get new server technologies approved within their IT deployment organizations.
We are excited that Java is joining the open source community and to contribute towards making Java a better home for dynamic languages.
Previously:
Sun's Open Source Strategy still MIA-An open letter to Jonathan Schwartz
Sun vs. Scripting Languages
Marketing to Dilbert: Mini-analysis of Sun
Sun coming out of death spiral-McNealy moving on?
Front End Integration-Lightweight Architecture Part 3
Enterprise SOA Apps Take Off on Lightweight Architecture
Google, Amazon, and Yahoo! point enterprise developers towards "lightweight" architecture
Posted by Dave Rosenberg on June 26, 2006 10:23 AM
May 23, 2006 | Comments: (0)
Does Ingres Matter? Conversation with Dave Dargo
I spent some time with Ingres CTO Dave Dargo today talking about Ingres and the market in general. Dave was at Oracle for many moons and is definitely one of the important database thinkers.
My initial conversation question to Dave was simple: Why does Ingres matter?
Dargo responded that he joined because Ingres had some good technology and a strong customer base. He also said that market-wise database customers were not terribly pleased with the existing products from Oracle and Microsoft.
"The success of Linux and Apache emboldened people to experiment with open source. Customers felt like they could go with free and not get features, or go with closed and be stuck. Ingres was already through the maturity process and had better technology than the other open source databases. I wanted to take Ingres beyond the database. If you look at MySQL and EnterpriseDB, they are consciously following the market. It's a valid model for them, but I wanted to do something Oracle couldn't do. Since there was already a good customer base, this was much easier for Ingres. Ingres can do things in the market that haven't been done before."
This last statement is interesting as it shows that the Ingres generic strategy is one of differentiation rather than the low-cost provider strategy often seen in open source.
One of the key pieces to new user adoption is the software appliance that Ingres is working on. Dargo cited this as an important development in that the appliance delivers a complete stack, tuned to run the Ingres database. The appliance prototype will be out in June and is expected to be in production this year.
Ingres is positioned and structured as a more traditional software company than say MySQL or EnterpriseDB. With over 10,000 customers (primarily from the CA spin-off) there is a sustained revenue stream. The business question is more related to where the growth comes from. I think the appliance idea is very interesting, I just don't know that anyone outside of the existing user base is looking to use the Ingres database. I also wonder if the execs embrace of the traditional software sales model is the right way to go. It makes sense in some ways, but also defeats some of the benefits of open source, namely lower COGS in relation to sales and marketing. Regardless, it will be interesting to see what happens.
Link: Dave Dargo's Blog
Posted by Dave Rosenberg on May 23, 2006 09:11 PM
May 14, 2006 | Comments: (0)
OpenClovis-Open Source Telecom Middleware
Here at the Open Sources Blog we are often lucky to get an early look at new open source companies and besides helping them get some press, we also inform the world of products that might be interesting.
Today we have an example of an audacious play backed by investment from AT&T and Intel as well as industry partnerships with HP, IBM and others. And it's in one of the most conservative IT markets of all -- telecommunications.
The company, OpenClovis (www.openclovis.com), is open sourcing a middleware HA and management layer of code that runs on Linux. The idea is that vendors write applications on top of this layer and hey, presto, instant HA and manageability. It apparently sits beneath the new telco platform stacks recently announced by Oracle and BEA.
OpenClovis is a bet on some big trends in telecommunications. When telco tanked in 2001, the shaken survivors of the industry began a race to escape the high costs of their broken business model. It started with the shift from proprietary, integrated hardware and software stacks. Smart vendors recognized that their competitive advantage and real value add were further up the solution stack. Plumbing was a commodity. That's when Linux began to gain traction on the software side. On the hardware side, Intel stepped forward with ATCA, and IBM with BladeCenter. These were all commodity volume opportunities for big vendors who know how to play this game to win.
While network equipment providers and telecommunication equipment manufacturers are re-examining the cost structures of their business models and embracing this trend to commercial off the shelf, wouldn't that be an excellent time to enter the market with open source middleware? Every telco application and service requires carrier grade-level availability and management. What if there were an open platform that provided that layer? Hey, no need for everyone to keep recreating that piece hardcoded into every new application. Vendors and carriers can focus on their real value add -- developing the actual service or application.
OpenClovis certainly has impressive industry backers. Now it's up to them to build the open source community and ecosystem they'll need to succeed.
Posted by Dave Rosenberg on May 14, 2006 11:19 PM
April 26, 2006 | Comments: (0)
SolidDB and MySQL making open source databases better for enterprise
I spoke with Paola Lubet, VP of Marketing at SolidDB to discuss their recent announcements with MySQL. According to Paola, the major roadblock for OSS databases into mission-critical environments is the real time nature of transactional data. SolidDB has been developing real-time databases for the last 14 years with features that go beyond basic storage (as one might perceive OSS databases to offer). This advanced functionality is common to Oracle etc, but not regularly seen in OSS databases. MySQL is the first open source DB that will feature this functionality. Both camps think that this will help MySQL move deeper into enterprises.
A prototype is available for download now.
Beta is July
Production is planned for Q4.
Posted by Dave Rosenberg on April 26, 2006 11:05 AM
April 12, 2006 | Comments: (0)
Quick Look: OpenQRM Systems Management
This morning I spoke with the exec team from Qlusters (FYI-it's pronounced Clusters, not Q-lusters) about OpenQRM, their open source system management platform.
The goal of OpenQRM is to provide a basis to start uniting other open source management tools in a unified matter that allows for each of use and deployment. The fundamental painpoint that led to the development of OpenQRM was the fact that IT shops were moving to commodity hardware but finding that management tools lacked-which defeated the very economics behind using cheap boxes. This sentiment is very similar to a number of other companies-mainly proprietary vendors who have built large scale apps (Cassatt), and appliances (Levanta) to manage the proliferation of Linux boxes and open source apps.
According to the team, OpenCRM has consistently ranked in the top 20 projects on Sourceforge.net for the last 3 months. There have been approximately 6500 downloads including the live CD which gives you a test drive of the full suite. There are somewhere between 200-500 installations that the company is aware of...possibly more out in the wild.
OpenQRM is interesting because it allows for integration with other applications, rather than forcing a data center change. Every IT shop has some level of monitoring in place so extending rather than replacing has an appeal. And for the moment they feel it's less about competition and more about plugging in various open source applications. Thus far the community has been responsive and supportive of the efforts.
Use Cases
New Data Center:
For new infrastructure things like provisioning are top of the list. The software allows for decoupling of images and policy driven management.
Existing Infrastructure:
If you already have a decent sized infrastructure, things like centralized management consoles have been very popular-aggregating all these disparate applications.
So how does Qlusters make money? There is the open source OpenQRM project that is not entirely supported. Qlusters provides a subscription to the code that is tested with paid support-similar to the RedHat model. In the pipeline is an enterprise version that will include features that will be exclusive to the paid support versions. List price is $750 per managed server, per year and is discounted at volume.
Posted by Dave Rosenberg on April 12, 2006 03:05 PM
April 09, 2006 | Comments: (0)
I'm in the process of evaluating a new CRM system to migrate off of our internally developed custom system and onto something that we don't have to maintain ourselves. In this final round I'm looking at Oracle, SugarCRM, NetSuite and Salesforce.com. The interesting (and depressing) fact is that any new application we consider requires a non-trivial level of development in order to integrate the systems together. In fact, it seems like engineering teams th