The State of Systems Management in 2010
2009 was a rebuilding year for the big four systems management vendors; BMC, CA, HP, and IBM. After the acquisitions of the last several years, the big 4 spent 2009 shoring up their monitoring and systems management frameworks to support the initiatives beginning to take shape such as IT Service Management, driven by the release of ITIL v3. Furthermore, concepts such as cloud-based computing and SaaS (software-as-a-service) highly dynamic infrastructures are starting to catch the eye of these vendors, especially in the marketing department. These vendors are all working diligently to reposition their solutions to support these virtual and highly dynamic computing platforms. Most of the acquisitions made by these vendors over the past several years have been to do just that. I last blogged about these vendors, and specifically application mapping, in 2008. Since that time, application mapping, and the emergence of a functional and useful CMDB are beginning to take shape with many of the Fortune 500. In fact, after attending the Gartner Data Center conference this past November, may of the Fortune 50 firms are undertaking some form of IT Service management effort, beginning with redeployment and scaled down approach to defining and implementing a CMDB. I thought I would take a few moments to discuss my view on this space for 2010.
Of the major vendors, Hewlett-Packard (HP) has one of the clearest messages and best-defined road map of how they are going to morph their solutions to operate in this changing IT landscape. The HP OpenView suite has a dizzying amount of products that are offered under the OpenView umbrella, but there are a core group of applications that when properly deployed, provide a compelling suite of tools to manage today’s dynamic computing platforms. HP OpenView Operations and HP OpenView Operations Orchestrator are two of those tools. Both tools utilize the discovery technology that has been a mainstay at HP for the last twenty years, Network Node Manager. The Achilles heal with this framework is the use of smart plug-ins, which are coded by HP. If you are a bleeding edge firm, then smart plug-ins will most likely be available for your technologies.
IBM still does not have a very clear message either for these dynamic environments. Some point solutions exist within the Tivoli suite, but they are point solutions for IBM specific environments and tend to be biased towards Websphere Application Server environments as well as Z OS solutions. IBM still does not have a clearly defined strategy of how their service management framework fits together. This lack of clarity is due largely in part to the fact that they have a gap when it comes to an incident and problem management ticketing solution. To fill this gap, IBM purchased MRO Software and their IT Asset Management and Service Desk Solution in August of 2006. Since the acquisition, IBM has struggled to differentiate themselves from the two vendors who have a large install base of incident and trouble ticketing systems; HP and BMC with Peregrine and Remedy respectively. Over the next 18 months, I believe IBM will do a better job of building a compelling IT Service Management suite around the MRO acquisition and provide better connectivity into many of the Tivoli network and infrastructure management solutions. Better integration into their application and dependency mapping solution is also in order, known as TADDM or Tivoli Application Discovery and Dependency Mapping
CA still has a disjointed solution, when it comes to monitoring high dynamic computing platforms based on a distributed systems platform. CA still provides a compelling suite of products when it comes to mainframe environments, which rivals those of IBM. CA has recently attempted to round out their offering in application performance management with the $375 million acquisition of Wily in January 2006. Ca has been adding additional functionality to the Wily set of tools, but the software and the individuals involve with it almost act as an independent company, and not fully integrated into CA. So when dealing with Wily, it often feels disjointed from the rest of the CA product suite. CA’s lacks some of the transaction tagging and per transaction performance management deep dive that one vendor, Optier, has built into their competing product, CoreFirst.
BMC Software continues to do a great job with articulating their value proposition and strategy when it comes to IT Service Management. BMC relies heavily on their customers that are loyal Remedy users to market and sell their other solutions to these customers, namely Performance Assurance as well as their Atrium CMDB solution.
BMC also completed the acquisition of Tideway Systems in October 2009. Tideway was the last independent vendor in the application discovery and dependency mapping space. Realizing that customers were having challenges with the deployment of the legacy BMC solution, foundation discovery and topology discovery, BMC made the acquisition to round out their strong capability to house a CMDB solution alongside a very robust workflow management solution, Remedy.
BMC has said that they intend to keep most of the Tideway Foundation product intact and continue to sell it as a stand-alone solution. Tideway already had an Atrium connector, so BMC will obviously make that connector much more robust and seamless. BMC has also mentioned that they will add network discovery capabilities as well as some level of mainframe discovery to the Tideway Foundation solution, which is now called BMC ADDM or BMC Application Discovery and Dependency Mapping.
For those looking to have better visibility into the actual performance of their applications and for great insight into where applications may be suffering from performance or other issues, Optier is a tool you should consider looking at. Although the initial implementation can be challenging and for most will require assistance from Optier Professional Services, the performance and availability information that comes from Optier is invaluable to determining where applications, especially multi-tier applications, are experiencing performance-related problems.
Optier can do this by “barcoding” each transaction that flows through the various tiers of the infrastructure. So for example, if you have a web-based application that is in a DMZ and utilizes an application server on one host, makes calls to a backend database server, and relies on an efficiently functioning web server and authentication infrastructure, Optier will provide round trip time and latency information across each of those tiers. It will provide information on where a transaction is taking the longest, so you can pinpoint performance issues very quickly with the Optier console.
If you have a latency sensitive application that is generating healthy revenues and margins for your firm, and you are having application performance issues, talk to the team at Optier.
Look for all of the vendors to further define their monitoring solutions to fit better with the cloud-based computing concepts as well as SaaS and IaaS environments. I expect that HP and BMC will continue to lead the pack in terms of having the best-defined strategy and more importantly the monitoring technologies behind the strategy. Look for BMC to release an updated BMC ADDM solution this fall, which will incorporate network and mainframe discovery into the solution. Also, look for very tight integration of the new BMC ADDM with their Atrium CMDB.
HP clearly leads the way in the application testing space, with their acquisition of Mercury several years ago. As expected, HP has fully integrated the Mercury CMDB solution into the OpenView product suite, and with tools such as OpenView Operations Orchestrator, it is quite a compelling solution. The combined solution is somewhat better than that offered by BMC since the instrumentation you do in the test and development environments, with tools such as LoadRunner, can be reused in the production environment, with tools such as HP Business Availability Center. The issue with these tools is they require a level of operational and process maturity that most firms do not possess, so the implementations fall short of expectations in many cases.