Infrastructure Management Software Blog : performance

Smart Decisions Today instead of Desperate Ones Tomorrow

pspielvogel — Thu, 05 Feb 2009 19:45:00 GMT

Yesterday, I had the opportunity to participate in a briefing in which a dozen IT executives from a leading financial institution came to discuss their unique requirements and how we can help them meet their aggressive growth goals. The first slide that their head of technology presented contained the equation “change = opportunity + risk”. This started a six hour discussion about market and technology disruptions and how they affect their quest for increased performance, capacity, and efficiency.

The theme of the day was business value. The customer had recently made several major changes to their IT infrastructure, all to embrace innovations that provided more performance for less money. Some of these upgrades meant dropping vendors of proprietary technologies who have served them well for years in favor of open platforms such as Linux. In financial markets, efficiency rules.

While efficiency was the foundation of many of their technology decisions, innovation was their passion. Advanced technology was the enabler that allows them to be first, best, and fastest in meeting their customers’ demanding needs. The company needs to develop new financial instruments to address rapidly changing and unprecedented (at least in recent history) market conditions. With rising transaction volumes (long-term trend, anyway) and rising customer expectations, there is a constant need to increase system performance and capacity. And, for IT solutions that can manage growing complexity.

In addition to technology changes, the regulatory environment for financial services firms is also shifting rapidly. While some decade-old regulations such as Glass-Steagall are now gone, new ones are taking their place and there will be new regulations to fix perceived free market inadequacies. This places additional load on IT systems as they must track and log every moving part to ensure compliance with new rules.

Reporting granularity is a major requirement for any IT system. While averages provide useful trend information, they fall short in deliverable actionable intelligence for troubleshooting. Generally, it is spikes that cause system outages rather than averages. And, if your data collection times are too broad, you lose the ability to focus on the indecent that caused an outage or lost data. This was a big discussion topic.

Another memorable take away from the day was when one of their executives said “I would rather make smart decisions today instead of desperate ones in the future.” This was the premise for setting up what turned out to be a very productive meeting.

What smart decisions are you making to simplify managing your IT infrastructure?

For HP Operations Center, Peter Spielvogel

ROI for IT Infrastructure Monitoring - Measuring what Matters

pspielvogel — Mon, 12 Jan 2009 18:30:00 GMT

I read an interesting post and related ebook by David Meerman Scott on why traditional marketing ROI measures lead to failure. His premise is that measuring marketing metrics such as number of sales leads captured and press mentions lead to the wrong behaviors and in some ways undermine one of the primary goals of marketing, which is to increase sales and market share. So, why is everyone in both marketing and IT so focused on ROI?

The answer is that focusing on the return of your investments in different parts of the business allows you to allocate scare resources and drive the best returns for the shareholders. They key is tracking the metrics that matter. Here are two examples, one from marketing and one from IT.

In previous positions, I have created marketing dashboards that charted many the items that Mr. Scott slammed. Why would I or other seasoned marketing professionals do this? One reason is that some metrics are relatively easy to track (such as leads captured at a trade show or the number of responses to a marketing campaign). Correlating these to the real goal of increasing sales is much trickier and requires much heavier monitoring infrastructure including obtaining accurate input from sales and customers about the number of touches and how individual marketing campaigns or programs influenced each stage of the sales process. Few companies have the will or discipline to do this.

On the IT side, there are also easily-trackable metrics. Server utilization, power consumption, and application uptime appear on many IT dashboards. While these are certainly important, what really matters is how the IT infrastructure supports the business goals. Business owners care about:

Availability - can my users access the applications they need?
Performance - does the application deliver an acceptable response time?
Data accuracy - does the application maintain data integrity?

Again, tracking these business-focused metrics is harder than focusing on ones that are easy to gather from element managers that often accompany systems. But, the right management software and some automated processes make it straight forward to create IT dashboards with mean. This is what the field of business service management is all about. BSM links the underlying infrastructure and applications to business outcomes such as those listed above.

To learn more about BSM, please visit my colleague Mike Shaw’s BSM blog or download a white paper about HP’s approach to BSM.

In future posts, I and my fellow bloggers will address how robust IT infrastructure monitoring contributes to delivering availability, performance, and accurate data

For HP Operations Center, Peter Spielvogel.