Killer server company? Are we still talking abt HP?

Hewlett-Packard recently solidified their enterprise software commitment with its intended $450 billion purchase of Mercury Interactive, a business software producer.

This is a big move for HP into enterprise software, and a strong sign of support from HP for application developers and testers. From a business perspective, an industry leader is taking a strong position on SOA and applying pressure to other pure play hardware, application, and services providers such as IBM, Oracle, and Dell. From a technology perspective, a unique opportunity exists to converge and offer software for end-to-end testing, network monitoring, application management, and governance.

This deal is largely concerning service-oriented architecture (SOA), an increasingly popular—perhaps the most popular—architectural model for enterprise software. It prescribes an approach where loosely-coupled, independent software agents interact using a standardized model.

So, what does this acquisition mean for developers and management staff responsible for SOAs and data center operations? First of all, it securely puts HP on the short list of every SOA architect, developer, and application manager. It also validates the need—and delivers the tools—for data center staff to control and assure the systems that drive the business.

The HP, Mercury, and Systinet combination provides a strong solution for governance, monitoring, and management of SOAs. And this is certainly good news for customers looking to embrace SOA to manage and simplify the complexity in their environments. It’s also a good first step to bridging the gap between operations and development.

So, what is missing? The story has mostly spun around SOA management, registries, and governance, tied, of course, to HP's core competencies in centralized IT operations and service management. Interestingly, testing was only briefly mentioned in the publicly released information by HP although more than 60 percent of Mercury's revenue comes from application testing tools. Moreover, this acquisition doesn't truly provide end to end testing of SOAs.

In this context, end to end SOA and integration testing means a few things:

1) Testing an entire business process path to assure that the integration has resulted in the intended execution of transactions, interactions, and data transformations

2) Testing across multiple platforms, transport protocols, enterprise service busses (ESB), language interfaces, and messages

3) Validating the linkages and integrations between business services and operational systems to meet target defect rates and SLAs

Increasingly today, companies are connecting to customers and suppliers and other third parties via disparate applications designed to work as a single business process through integration and SOA. Without testing at the middleware, ESB, protocol, and message level, how can HP be sure that a SOA and supporting composite applications work as designed? How will HP ensure service level agreements of SOAs without isolated testing and visibility into the variety of moving parts? And what good is a registry of inter-connected and inter-dependent services if they aren't tested end-to-end? This "behind the screens" testing is needed to maintain the quality of integrated applications, assure service levels, and deliver agility to the business.

The question remains: will HP provide solutions for testing and validation of everything that is happening underneath the SOA or "behind the screens"? It will be interesting to see if HP takes the next step in service management and delivery by combining network monitoring, registries, governance, and application testing tools, with true SOA and integration testing.

The news is encouraging for HP, Mercury, and for end-users of each of these companies. HP and Mercury now provide a strong solution for application, SOA, and network governance and management. It is also encouraging news for vendors like Solstice Software who provide incremental value to HP/Mercury customers via an end to end SOA and integration testing suite.

Chris Benedetto
Vice President of Marketing, Solstice Software
www.solsticesoftware.com

Last week I posted an open letter with a perspective on why the HP acquisition of Mercury Interactive is not in the best interest of CIOs and IT organizations. Over the past few years we in the IT Performance Management market have seen a continued concentration of portfolio management solutions within the large IT infrastructure software providers, HP acquisition of Mercury is yet another example of this trend.

One of the major complaints about IT organizations is their lack of alignment with the goals of the Business and the lack of transparency in the technology decision-making process. Successful CIOs understand that they need to be an innovation partner with the Business and communicate the value of technology in business terms. These IT leaders align themselves with the Business Demand in comparison to the IT Supply.

The IT Infrastructure supermarkets "HP, CA, IBM, etc." have spent decades developing products to support the IT supply-side. This is their #1 focus, and it will remain so indefinitely. In the case of Mercury, the testing business is clearly a supply-side business, while Mercury’s portfolio management offering is on the demand-side.

With every portfolio management acquisition, the pressing need for demand-side solutions is getting lost in a sea of supply-side tools.

The result of all this is that at a time when alignment with IT demand is the #1 priority for the CIO, Supply-side vendors are diluting the viability of the demand-side offerings by burying them within their classic IT-operation-centric product lines.

A better approach for the CIO is to create a clear separation between Business Demand and IT Supply, and to then optimize the organization’s demand-side capabilities (people, processes, and tools). Demand after all drives supply, and not the other way around.

Maintaining a clear demand-side focus allows CIOs to:

- Ensure the best demand-side processes and solutions are established. This is the strategic imperative and should not be bundled into a supply-side contract renewal as some kind of afterthought.

- Avoid even the impression of a bias to internally supplied IT services managed by the IT supply-side tools.

- Partner with solution providers uniquely qualified and focused on the demand-side challenge. This challenge is a business-centric issue, not an IT operations management issue.

- Ensure a healthy vendor portfolio. Although it is tempting to one-stop-shop, this approach eventually consolidates power with vendors and limits customers’ ability to innovate.

- Also remember that there are significant technological developments on the near-term horizon that question the need for the single vendor, integrated suite (SOA, Web Services, etc.).

Clearly, as the previous entry states, the Mercury monitoring tools are a no-brainer for HP to integrate. They are a natural extension of the OpenView systems management product line. But will the portfolio management solution fare as well? Again, as the previous entry points out, does HP have the domain expertise and services strength to truly deliver this solution to the CIO? Ultimately, will the demand-side get lost once again in the supply side?