Log Management for IT Ops: 5 Best Practices
February 23, 2015

Jim Frey
Kentik

Share this

Log data may be many things, but one thing is for sure – it isn't sexy. In fact, in most cases, it's downright ugly, because there are really no standards out there for how log data should be structured. For decades, this fact has kept log data from being a practical source of information for anything beyond a few specific use cases, such as watching for important events (like system reboots or config changes), security monitoring (like firewall blockages), or deep troubleshooting.

Times have changed, and the most recent crop of log management vendors have taken advantage of the steady growth in processor capacity to overcome the complexity and scale challenges of harvesting and analyzing all of the log data that an IT infrastructure continuously throws off. Now there are practical ways for taking advantage of the unique perspective and insights that log data can provide on a much broader basis.

In my last post, I shared some key findings from an EMA research report published last fall that dove into the ways in which log analytics is being used to support network operations. Building on that, following are a couple of the recommendations that EMA is making on how best to think about log data as part of an integrated management architecture and strategy:

1. Think twice before planning to store all log data

While most organizations are gathering log data for analysis on a continuous, ongoing basis, only a third are storing all log entries all the time. Interesting, those organizations considering log data to be "strategic" are actually much less likely to be storing all log entries all the time than those who consider log data to be "tactical". Strategic log users prefer instead to be more surgical, looking for specific types of logs or storing all log data only when certain trigger situations occur.

2. Consolidate your log analysis tools

We find that an overwhelming majority of organizations are either currently using one centralized log analysis system or are planning to consolidate the multiple tools that they have into a single system. This makes tremendous sense if you are trying to get the most out of your log data either in support of integrated operations or simply for better collaboration and cross-team sharing.

3. Focus on fast and intuitive search capabilities

The number one challenge voiced with respect to analyzing log data is knowing what to look for. It's not surprising then that the most popular feature that IT pros look for in a log data analysis solution is fast search. The latest generation of tools have made quick and effective search a high priority, and if you don't have such capabilities in your current system, you should consider an upgrade or alternative.

4. Don't implement log data analysis as an island

Consistently, we find that organizations are getting the most value when log data collection and analysis is integrated with other data sets and analysis systems. This can be done either via log data collection/analysis tools incorporating non-log data themselves or by openly sharing log data with other management aggregation systems. Some of the strongest values are being achieved by connecting the insights available from streaming log data with other performance monitoring measures, to proactively recognize performance degradations and related root causes.

5. Log data is relevant for BSM/ITSM

EMA has found a very high usage rate of network log data for higher level BSM and ITSM type initiatives, such as service quality monitoring, unified IT operations, and CMDB. Such usages were particularly high among those who consider log data to be strategic rather than tactical. So even though log data may be ugly, don't overlook its importance in supporting your highest level management objectives.

There were a couple of surprising dichotomies uncovered in the research study as well. For instance, the top reason people value log data is that they consider it to be cost-effective, however the second greatest challenge was identified as cost of tools. Another example involves just how effective log data is. The second highest perceived value was faster time to resolution than other data sources, however the number one challenge was knowing what to look for.

Clearly there is great and growing value in collecting and analyzing log data for IT planning, operations, and security. And while there are still challenges to be faced, best practices are emerging to help everyone understand what to expect and how to get the most returns on investments into log data collection and analysis tools.

Jim Frey is VP of Strategic Alliances at Kentik
Share this

The Latest

November 17, 2017

Just in time for the holiday shopping season, APMdigest asked experts from across the industry for their opinions on the best way to measure eCommerce performance, in terms of applications, networks and infrastructure. Part 3, the final installment, covers the customer journey ...

November 16, 2017

Just in time for the holiday shopping season, APMdigest asked experts from across the industry for their opinions on the best way to measure eCommerce performance, in terms of applications, networks and infrastructure. Part 2 covers APM and monitoring ...

November 15, 2017

As the holiday shopping season looms ahead, and online sales are positioned to challenge or even beat in-store purchases, eCommerce is on the minds of many decision makers. To help organizations decide how to gauge their eCommerce success, APMdigest compiled a list of expert opinions on the best way to measure eCommerce performance ...

November 14, 2017

More than 90 percent of respondents are concerned about data and application security in public clouds while nearly 60 percent of respondents reported that public cloud environments make it more difficult to obtain visibility into data traffic, according to a new Cloud Security survey ...

November 13, 2017

Today's technology advances have enabled end-users to operate more efficiently, and for businesses to more easily interact with customers and gather and store huge amounts of data that previously would be impossible to collect. In kind, IT departments can also collect valuable telemetry from their distributed enterprise devices to allow for many of the same benefits. But now that all this data is within reach, how can organizations make sense of it all? ...

November 09, 2017

CIOs trying to lead digital transformation at the speed needed to succeed need a mix of three scale accelerators, according to Gartner, Inc. The three scale accelerators include: digital dexterity, network effect technologies, and an industrialized digital platform ...

November 08, 2017

While the majority of IT practitioners in the UK believe their organization is equipped to support digital services, over half of them also say they face consumer-impacting incidents at least one or more times a week, sometimes costing their organizations millions in lost revenue for every hour that an application is down, according to PagerDuty's State of Digital Operations Report: United Kingdom ...

November 07, 2017

Today's IT is under considerable pressure to remain agile, responsive and scalable to meet the changing needs of business. IT infrastructure can't become a bottleneck, it must be the enabler. But as new paradigms, such as DevOps, are adopted, data center complexity increases and infrastructure constraints can block the ability to achieve these goals ...

November 06, 2017

It's 3:47am. You and the rest of the Ops team have been summoned from your peaceful slumber to mitigate an application delivery outage. Your mind races as you switch to problem solving mode. It's time to start thinking about how to make this mitigation FUN! ...

November 03, 2017

With the increased complexity of IT environments, the rising cyber threats and the growing number of IT alerts, IT organizations have come to the realization that throwing more people at IT issues doesn't solve the problem ...