Performance Critical App News

OK – I don’t think you are stupid (my editor thought the title was catchy). But we’ve all done things that, in retrospect, seemed pretty backward because we were unaware of a better approach.  Traditional approaches to managing application or IT infrastructure performance is a case in point...

Anomaly Detective detects application issues as they develop, even across multiple sourcetypes, and quickly diagnoses them. Anomaly Detective installs as a Splunk app and is operational in minutes – no data export or import required...

According to a recent Aberdeen survey, organizations with real-time performance management tools are four times more likely to detect the root cause of an application issue than those without. Eighteen percent of organizations with real-time performance management are able to reach 99.996 percent or greater error free transactions compared to only eight percent of organizations lacking this capability. Fifty-four percent of real-time enabled organizations have 90 percent transaction success compared to only 30 percent of those not enabled...

CA World kicked off in Las Vegas and was all a buzz as CA announced the availability of Application Behavior Analytics with CA APM 9.5 (courtesy of our previously announced CA / Prelert OEM agreement). In their announcement, CA says that this new analytics capability helps eliminate problem resolution guesswork while making it faster...

Cross silo correlation is hard-to-impossible for many IT organizations, but it’s critical for discovering anomalies within composite applications and their relationships with each other and other aspects of the IT infrastructure. And finding anomalies across related systems is often the key to troubleshooting and repairing application performance issues...

The truth is that KPI’s and thresholds have never been completely reliable performance indicators but they were the best tools available and worked pretty well in the more well defined physical, client/server IT environments of yore. Unfortunately the Web, the cloud, virtualization, and agile development have caught up with IT and today these indicators are verging on useless...

The standard way to build a Splunk custom search command is to write a Python script that receives data in CSV format via its standard input , and outputs its results in CSV format via its standard output . In terms of simplicity this is great, but what about performance?

KPI’s and thresholds have their place in IT infrastructure management, but as your environment becomes more complex they become less and less informative. In a recent blog, “Why Anomalies”, we made the point that excessive IT focus on KPI’s and thresholds can hide a lot of IT infrastructure issues. Here are some examples...

Both Java and C have well known APIs for manipulating TCP socket connections. When you read from a TCP socket into a buffer in Java, you call the read() method of the Reader class.  In C you call the recv() function (or less portably the read() function).  In both languages these calls usually return the [...]

Today, we announced the addition of a new reseller to our portfolio of global channel partners. Kiratech Srl., an information technology and communications solution provider based in Italy, has signed a reseller agreement with us to offer Anomaly Detective for Splunk Enterprise environments. Kiratech has decades of experience as an IT solutions provider in the [...]

There are probably two reasons why most people responsible for assuring or restoring IT infrastructure or application performance haven’t already thought, ‘what I need is tool that automatically detects anomalies’. First, when you are spending too much time diagnosing issues you may not even have anticipated...

Machine learning is a branch of artificial intelligence that employs a variety of techniques and advanced algorithms to uncover information hidden in vast quantities of data many times faster than a human ever could. The technology is the basis for many important IT related functions, including spam identification and optical character recognition...

If you’ve read the blog posts of my colleagues, you’ll have learnt about how Prelert’s technology can quickly diagnose application outages and performance issues, and how Prelert is growing as we sign OEM agreements and bring on-board partners to resell our software.  However, over the coming months I’m going to post some articles about Prelert coming from a completely different angle, namely how the products are built...

Today we introduced a new reseller to our partner base — ICS GmbH. ICS, an innovative technology provider for IT service management, will offer Anomaly Detective for Splunk Enterprise environments to its base of customers, which includes the top 500 companies and public agencies in Germany, Austria and Switzerland. Our partnership with ICS will allow [...]

This week we announced the addition of another reseller to our growing list of channel partners. Function1, a leading enterprise software technology provider with offices in NYC, DC and Toronto, now include Anomaly Detective in their product and service offerings. This is exciting because Function1 understands the challenges that companies face with managing the huge...

You’ve probably read about using predictive analytics to analyze and predict customer behavior in order to increase clicks, conversions, and purchases. The same predictive analytics can be used to address mountains of machine log and management data as well. If you work for IT you no doubt wish you could spend less time troubleshooting application outages and performance issues and more time harnessing IT to provide business value...

According to a recent IBM survey, 78 percent of respondents work for organizations that link IT and reputational risk directly and those companies do tend to have better reputations. As one respondent puts it, “IT…. Is like the heart pumping blood to the whole body, so any failure could threaten the whole organization’s survival.” Forty-one percent said brand risk was most strongly affected by IT risk. Forty-six percent linked customer satisfaction directly with IT risk...

Today Prelert had the pleasure of announcing an OEM agreement with CA Technologies, a market leader in infrastructure and application performance management. CA Technologies will embed Prelert's anomaly detection analytics in CA APM...

John Michelsen is the chief technology officer of CA Technologies and is considered one of the great innovators in business IT.  With over 12 patents awarded to him, John talks about trends in the business IT world and how it has developed over the years...

You can’t consume the cloud like you consumed traditional IT. Cloud economics only make sense when you use only what you need just when you need it and when you only pay for what you need. That applies to performance management tools as well...

Big Data is not just a tool for business analysis. it can also be a helpful tool to improve the data center...

The ‘Design for Failure’ approach differs in that an application will be designed to ‘recover’ from any performance challenges without any reliance on the underlying ‘local’ infrastructure...

CA Application Performance Management (APM) 9.5 contains Application Behavior Analytics which automatically and algorithmically turns application performance metrics into proactive anomaly detection. “CA APM’s ability to mine huge amounts of performance data to identify potential issues before they become major problems is one of the many ways it delivers reliable business services that provide the best possible end-user experience...

If you're looking for a U.S.-based cloud provider to host your business applications and data, you might want to take note of one new report that ranks the performance levels for providers in 2012. The results might surprise you...

Gmail, Drive and several other Google applications were affected for several hours this week. A malfunctioning log-in system affected millions.  At its peak, 50 percent of the Admin panel and 60 percent of Google Sync login requests prevented access to a variety of Google applications on Wednesday, including Gmail and Drive. The improper configuration, introduced on Tuesday, caused log-in requests to be funneled to a small number of servers, which in turn ran out of capacity, and the overload caused them to malfunction...

In this post we provide 3 insights to testing applications that live, in any part, in the cloud.  The key takeaway is this:  Performance testing and tuning of the end user experience needs to be the focus of cloud application performance management.   Is the experience being delivered going to delight your end users?

When the service went down last week Google Drive users learned the hard way that it can be very risky to store their important files exclusively in the cloud. . While the outage lasted no more than a few hours for most users (some were back online within 90 minutes), it was another painful reminder of the risk involved in storing files online...

Apple outages have been occurring quite a bit recently, with iCloud getting the brunt of most of the outages, but iMessage and FaceTime have both gone out in the recent past. We’re obviously not sure why these outages are occurring more often than in the past, but it seems users are starting to get fed up with it...

When the cloud crashes, there is an outage of the infrastructure, servers, or services at the core of the hosted service and the service comes down, or “crashes.”

Network and application performance optimization will no doubt become one of the fundamental goals for proactive organizations in the future. Cloud technology may still need refining, perfecting, but, like the Internet, it is here to stay. More importantly, it will open new avenues of opportunity, discovery and productivity that have, till now, only been imagined...