Aug 28, 2012 (08:08 AM EDT)
Splunk Adds Cloud-Based Systems Monitoring Option
Read the Original Article at InformationWeek
Systems monitoring and data-analysis specialist Splunk on Tuesday made its cloud-based Splunk Storm service generally available after months of beta testing. The service is aimed at companies developing and running applications on public cloud platforms such as Amazon Web Services (AWS), Microsoft Azure, Rackspace, and Salesforce.com's Heroku.
The Splunk Storm service runs on AWS, but companies can use it to monitor applications that might be running on a mix of platforms. Those apps might also be written in a variety of languages, including Ruby, Java, Python, PHP, and .Net. The idea is to offer a cloud-based option for monitoring and troubleshooting applications and gaining insight into system performance and downstream business conditions.
"Companies that are going cloud-only don't want to have to download software and run it in a data center, so we're offering Splunk as a service to help developers and startups that want to do it all in the cloud," said Leena Joshi, senior director, solutions marketing at Splunk, in an interview with InformationWeek.
Splunk's software has served much the same role in on-premises deployments, where the software's ability to monitor and analyze complex and variable machine data at high scale has secured the company's place in the big-data arena. The big data twist also contributed to the company's successful IPO in April.
[ Want more on Splunk's on-premises capabilities? Read Splunk Answers Business Demand For Big Data Analysis. ]
Splunk Storm is decidedly not a big data play. The on-demand software can still handle complex and variable machine data, but the service starts at $20 per month for 2 gigabytes of storage and tops out with a $3,000-per-month plan at 1 terabyte of storage. Companies looking for higher scale will typically turn to on-premises deployments, according to Joshi.
Splunk Storm customers can send data to Splunk Storm through streaming protocols such as TCP or UDP. Files can also be uploaded directly or sent through Splunk forwarders installed on servers and set to monitor specific log files.
Splunk Storm will be particularly appealing to organizations that need to add monitoring and troubleshooting capabilities to custom-built cloud-based apps. Packaged cloud-based apps, such as those from NetSuite, Salesforce.com, SAP SuccessFactors, and Workday, tend to have their own administrative and monitoring tools. Joshi says Splunk's on-premises software is more often used to monitor software as a service, but it depends on the service's openness with system performance data.
"These providers don't always want to tell customers about their service levels, but when they do have mature APIs available, we can put that data in Splunk and use it for system performance analysis," she said. As an example, Saleforce.com has an API that Splunk can use to monitor that application alongside other applications, whether they're on-premises or in the cloud.
Splunk said hundreds of beta customers are already using Storm, which is accessed online at Splunkstorm.com. A free tier of the service stores up to 1 gigabyte of data. Customers can scale their storage needs as required, but the pricing plans are based on monthly commitments rather than per-usage charges.
Even small IT shops can now afford thin provisioning, performance acceleration, replication, and other features to boost utilization and improve disaster recovery. Also in the new, all-digital Store More special issue of InformationWeek SMB: Don't be fooled by the Oracle's recent Xsigo buy. (Free registration required.)