eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
Oracle on Monday plans to release an update to Enterprise Manager that slaps at two pain points in grid computing: getting IT services available to meet agreed-upon service levels, and being able to measure whether IT is helping the business out or dragging it down.
With Oracle Enterprise Manager 10g R2, Oracle Corp. aims to help IT administrators figure out whether IT services are actually available and meeting SLAs (service-level agreements) by plugging service dashboards into Oracle Grid Control, which is the flagship product in EM 10g.
The Grid Control update also features service modeling and graphical service topologies that display both historical and real-time metrics, along with service levels impact on business performance.
Mary Johnston Turner, an analyst with Summit Strategies, said Oracle is doing exactly what it needs to in order to respond to the migraines of grid.
Turner said that her firm has recently finished a survey looking at awareness of, adoption of and pain points resulting from virtualization, grid and SOA (Service-Oriented Architecture).
The research found that those technologies are giving administrators headaches particularly when it comes to figuring out how to judge or prioritize their IT investments.
“When we talk to folks, they tell us theyre trying to do [a few] things: They talk about service, availability and uptime, and business process performance,” she said.
“When we talk to enterprises about how do you really judge or prioritize your IT investments, those are two things they talk about the most: Are IT services available [and] meeting SLAs, and, Are we making a contribution to the performance of the business?”
Businesses dont want to know just whats going on with individual components of a virtualized computing scheme, Turner said. What they want to understand is whats going on end to end.
To do that, they need a view of root cause, they need impact analysis and they need dashboards—all components of the EM upgrade.
The new Root Cause Analysis feature is designed to help administrators to quickly pinpoint and understand the impact of problems with performance, availability and capacity throughout the application infrastructure stack.
Service and system dashboards are meant to give a single view of the overall status—availability, performance and load—of key business services and applications, as well as the status of the systems and infrastructure underneath those applications.
In addition, new IT compliance and configuration policies help administrators to monitor configuration and performance.
Oracle also beefed up automation in EM R2. Jay Rossiter, vice president of systems management products, said the focus was to reduce the labor involved in deploying grids, and to reduce the errors that result from manual operations.
Oracle particularly zeroed in on the patching process, where manual errors are particularly hazardous.
With the update, RAC (Real Application Clusters) provisioning will let administrators convert a sole Oracle Database into a clustered database with RAC. New clusters can now be deployed automatically.
“We can provision multimode RAC systems, extend them, add more processors or Blade servers and have them be clusterized,” Rossiter said.
Oracles also out to gut the grunge from Linux system configuration, Rossiter said.
“[Youll be] able to image Linux operating systems, and with software thats been pre-tested on it,” he said—all a testament to the popularity of Linux running on Blade systems beneath grid.
There hasnt been much capability to support Linux in such an environment up to now, Rossiter said—a state Oracle is seeking to change in EM R2.
“We have focused on this capability around imaging Linux systems because we want to enable customers to roll in new Blade servers and take pre-tested software and get it up and running in an automated manner.”
With the update, Linux and grid will be a much easier marriage to forge, he said.
“Imagine we have [taken] the pre-tested bundle, stamped it down and performed all the grungy steps needed to get the system up and running,” he said.
“With the RAC system, that means being able to put all the pre-tested pieces of software on the box. In the case of RAC, dealing with all configuration issues being involved in having a cluster.”
As far as patching goes, new reference-based patching will provide what Oracle says is a “lights out” solution to patching Linux servers—the biggest barrier to rapid application deployment, Oracle claims.
EM R2 relies on a configuration capability wherein Oracle can inventory hardware, software and configurations into a single database.
When Oracle posts a critical patch, it can notify customers who are running EM when the patches are available, can highlight the patch to people administering the systems, and can identify systems that need the patch, can download the patch and can automate the patching process, Rossiter said.
Oracle does, in fact, need some help with its patching utility.
Next Generation Security Software Ltd. in August reported that Oracles Opatch utility was misleading enterprises as to their patch status, neglecting to inform them of necessary post-installation tasks including updating components such as PL/SQL packages and Java Class files in the database.
If these post-installation steps arent taken, the server remains vulnerable in spite of the Opatch utility having indicated that the server is in fact patched.
Rossiter wasnt immediately able to say whether EM R2 will address those problems.
Oracle EM 10g will be generally available on Monday on all supported Linux platforms.
New management packs include Configuration Management pack for Non-Oracle Systems, for $3,000 per processor, and the Provisioning Pack, also for $3,000 per processor. A Service Level Management Pack is available for $3,000 per beacon and $100 per test.
Oracle is also rolling out System Management Plug-Ins to support partners products. Hosts are $1,500 per processor. Non-Oracle middleware plug-ins cost $1,500 per processor. Network Devices costs $1,500 per Network Device, and storage costs $1,500 per terabyte.