One of the hottest topics in the data center lately is around big data and the actual dollar value that businesses are deriving from making sense from tons of unstructured data. Virtually every field is turning to gathering big data, with mobile sensor networks, cameras everywhere, and information archives. New techniques are being developed that can mine vast stores of data to inform decision making in ways that were previously unimagined. The fact that we can derive more knowledge by recognizing correlations can inform and enrich numerous aspects of every day life.
Cisco is partnering with leading software providers to offer a comprehensive infrastructure and management solution, based on Cisco Unified Computing System (UCS), to support our customers’ big data initiatives. Taking advantage of Cisco UCS’s Fabric based infrastructure, Cisco can apply significant advantage to big data workloads.
There are actually many advantages to hosting big data applications on Cisco UCS infrastructure. With UCS, Cisco offers a balance of performance, management and scale that sets UCS apart from other industry solutions. Although we’ll be discussing the benefits in more detail at Cisco Live next week, here is a sneak peak of what you can expect:
Reason #1 to deploy Cisco UCS for your big data analytics: Form factor independence and administrative parity.
Cisco UCS provides a single point of management for the overall infrastructure—whether it’s blade architecture on the enterprise application side or rack architecture on the big data side, including troubleshooting, monitoring, and alerting capabilities. Customers can proactively monitor the system and keep operational costs down.
In other words, Cisco UCS Rack Servers can be managed the same way as UCS Blade servers with full workload mobility across both blades and racks. This simplifies the management construct and eliminates the need for additional management silos in the data center. This form factor independence is made possible by Cisco Unified Fabric with single wire management and Cisco Unified Management that includes UCS Manager with Service Profiles.
Reason #2 to deploy Cisco UCS for your big data analytics: Consistent and Rapid Deployments using UCS Service Profiles.
Cisco UCS is a stateless computing architecture that relies on a service profile construct that abstracts a complete hardware state. Cisco UCS service profiles are a powerful tool for streamlining the management of modern data centers. They provide a mechanism for rapidly provisioning servers and their associated network and storage connections with consistency in all details of the environment.
The service profile encapsulates all the configuration aspects of an application, workload and bare-metal infrastructure.
During deployment, customers define their hardware profile in a service profile template and apply it to each node in the cluster. This enables customers to deploy multiple nodes in literally minutes. To scale, customers use the same service profile template for new nodes, eliminating the need to provision each server independently. The great benefit here is agility and faster time to market.
Reason #3 to deploy Cisco UCS for your big data analytics: Massive Scale with Cisco UCS Central to 10,000 Nodes.
As organizations scale their Cisco UCS infrastructure, managing it consistently with global policies is ever more important. Cisco UCS Central manages multiple Cisco UCS instances (or domains) across globally distributed data centers, allowing IT departments to enjoy the benefits of simplified, unified management of compute, network, and storage access resources.
Within a single UCS domain, Cisco UCS Manager can manage up to 160 servers or 10 racks of 16 UCS C240 M3 Rack Servers. Cisco UCS Central can provide global definition capabilities for policies and resource pools which can be flexibly allocated across distributed data centers. This enables administrators to follow a “define once, deploy many times” workflow for their compute infrastructure and provides the ability to manage up to 10,000 nodes. This becomes a unique and valuable feature for large big data deployments.
Have you considered Cisco UCS for your Big Data projects? I’d like to invite you to come and hear more next week at Cisco Live in Orlando. We’ll have a number or demos and experts on hand to answer all of your questions.
Stop by and say hello and let me know if you have any comments or questions, or via twitter at @CicconeScott.
Definately seems CISCO will take over the market on big data projects as I have yet to see If Oracle or any other`s to show up with the same kind of approach.
Thanks for your feedback Rob, we’re optimistic about UCS on Big Data but always great to hear it!
Much appreciated,
Scott
Hi John,
Thanks for your question. Cisco is working with a number of leading ISV providers to bring big data solutions to market. The three types of big data analytics solutions that Cisco supports are Hadoop, NoSQL, and MPP databases. However, the majority of deployments we are seeing currently are Hadoop deployments from a variety of ISV’s. Some of the more popular distributions of Hadoop that we support include Cloudera, Hortonworks, Intel, and MapR. However there are a number of other distributions that Cisco supports with our Cisco Common Platform Architecture (CPA). For a more complete list and to find the most recent solution briefs, you can take a look at our Cisco Big Data web page at: http://www.cisco.com/go/bigdata
Regarding Oracle, they do offer a release of Oracle NoSQL that Cisco also supports with Cisco CPA that has gained some traction in the market.
Thanks again for your comment and I hope this was helpful.
Best Regards,
Scott
Great hardware but who are the software ISVs partners to deliver big data apps cause it ain’t gonna be Oracle….