Easy Cloud Management: Large Data analysis SaaS

When it comes to managing local systems, both virtualization and cloud computing departments want to respond quickly to complex queries without having to deal with the underlying http://www.aliyun.com/zixun/aggregation/13748.html "> Infrastructure." Some of these sectors have increased confidence, using emerging products, such as "Big Data" analytics, delivered via software as a service (SaaS).

"They can simplify very complex information into views and put them on a page that anyone can view," says Nathan Smith, a senior engineer at centered NX, a San Francisco desktop service provider.

Smith has experimented with Cloudphysics, a cloud-based test product that uses large data technology to synthesize data. The aim is to compare data across different customers to provide a deeper perspective of virtualization and cloud infrastructure, and to leverage resources from VMware's vcenter statistics, rather than machine logs. "There is no need to be an expert because it has been decomposed into very simple forms," Smith said.

Centered NX hopes to make it easier to find fault storage performance problems in the VMware infrastructure and to speed up its customer service response by centered NX's large data analysis software, a service, Smith said.

Large data analysis software is how the service works

Other companies provide large data analysis software that includes sumo Logic, Splunk, Appfirst and Scalextreme.

Limelight NX is a web presence management company that supports customers through its own SaaS offerings, with a pool of most 600 physical servers behind it. In this infrastructure, the Web content management business accounts for about half of the servers. After trying to synthesize its local machine log failure, the company became the early adopters of sumo logic services.

"I can type in queries and provide data to me through two different data centers and different applications, which are essentially real time," Tom Cignarella, a Limelight NX's SaaS Technology VP. Limelight has so far deployed sumo logic to the server on its web content management business.

"We've been having problems with disk space, how to transmit all of these headaches," Cignarella said.

Sumo Logic was officially released in January, integrating machine logs into the cloud, and customers can query them. It uses components such as Hadoop Distributed file systems and Cassandra, as well as its own distributed database, Bruno. In addition, Sumo logic uses dynamic algorithms (patent-pending algorithm) to reduce the 1 million log information to 10-15 categories so that customers can find it faster in their cloud infrastructure. "It's better for us to work," Cignarella says. "We can dig through personal log files, but it's easier to narrow down the scope." ”

Large data analysis software as a service is still developing

Some of the features of Sumo logic are still evolving, just as it is a very young company. For example, Cignarella and his people can search queries on comprehensive log data, but the control Panel tracks the most common queries. Limelight Log and other sumo customer comparison analysis, this is the sumo of large data analysis on the other hand, is also cignarella can not see. Another common feature of large data analysis, anomaly detection is also to be part of this product.

Sumo Logic officially declined to comment on the performance of products that have not yet been released. Cloudphysics has a long way to go before it finishes its product, Smith says.

Cloudphysics's software summarizes its individual customer environment, presenting it in a smooth interface, with each unfolded graph (called a "card") showing information about different aspects of the infrastructure. Centered NX has helped Cloudphysics develop its storage performance cards, for example, in the chat process, new cards will appear.

"Now we are working on a card that is changing the ha configuration, so like an emulator ... will show how your host is being affected, "he said.

(Responsible editor: Lu Guang)