The Lowly Rack Rises to New Heights in Cloud Data Centers
Nobody thought much about racks.
It wasn't all that long ago that data center racks-those industry-standard, 19-inch cabinets-were largely an afterthought in IT environments. Nobody thought much about racks. They were basically dressed up metal frames that played a minor role to support the real star of the data center show: the server.
Today, all of that has changed. In the new world of cloud computing and big data, there is an intense focus on the rack. Engineers now think through every element of the rack, and in many ways the rack has become the new unit of compute.
This shift in focus stems from the need to achieve ever-greater densities and efficiencies across the data center. In these efforts, engineers design racks to accommodate the ideal mixes of servers, storage devices, and networking gear for targeted applications. They further enrich the rack with features such as power management capabilities and in-the-rack battery backup. And while they are at it, they optimize the entire rack for the challenges of cloud and big data.
This evolution of the rack revolves around three overarching needs in cloud data centers: efficiency, computation, and storage.
At the efficiency level, engineers now develop platforms that are optimized from the outset to meet cloud and big-data needs. In these efforts, they look for technologies that allow them to increase densities within the rack, such the two-socket servers in the Intel® Xeon® processor E5 family and single-socket Intel® Xeon® E3 1200-v2 micro servers.
At the computational level, engineers are designing and optimizing racks to the deliver the highest level of performance to handle the workloads of big data and cloud services.
And at the storage level, engineers are looking to improve storage economies and performance by scaling out storage and moving the disk closer to the compute-which increasingly means putting the storage right in the rack, rather than in a standalone array.
The components that go into the rack are selected and optimized for the needs of targeted applications. The entire platform is developed to meet cloud and big data needs. The goal is always to deliver the best performance for applications at the lowest possible cost.
These are all the realities of the world of cloud and big data. And in this new world, everything revolves around the rack.
Dylan Larson is the Director of Server Platform Marketing at Intel. Dylan has 10 years of experience with Intel bringing new technologies to market.
The above insights were provided to InformationWeek by Intel Corporation as part of a sponsored content program. The information and opinions expressed in this content are those of Intel Corporation and its partners and not InformationWeek or its parent, UBM TechWeb.
Multicloud Infrastructure & Application ManagementEnterprise cloud adoption has evolved to the point where hybrid public/private cloud designs and use of multiple providers is common. Who among us has mastered provisioning resources in different clouds; allocating the right resources to each application; assigning applications to the "best" cloud provider based on performance or reliability requirements.