Application Aware DataCenter Infrastructure

Next-Gen Data Centers

The basic idea is to embed functions currently done in middleware server software into router silicon, thereby gaining speed and scaleability and, much more fundamentally, giving the router sight of what it is actually shifting around the network, and therefore being able to route it according to what it contains. In principle, for example, this means that service providers could offer different QOS to reflect the importance of the traffic content (so bills and tickers get through fast, and illegal downloads get lost), and multicast content push services could be handled directly by the network, rather than by server farms.
This enables the network to classify and prioritize content based on specific application-layer attributes – for example, the dollar value of a purchases order. Application-aware networking devices can also communicate this classification to the underlying IP network via mechanisms like DiffServ code points or 802.1p bits. For example, carriers demand very high service availability, and therefore a networking product must have high-availability features such as active/active failover. Products also need to be able to interwork with the existing infrastructure, which typically will include MPLS VPNs supporting multiple enterprise networks. And there must be some way for service providers to monetize the flow of message traffic through application-layer billing capabilities.

Content-based QOS

As the issue of power and cooling heats up, companies must carefully evaluate their data center planning. A simple cooling unit and a raised floor just would not cut it anymore. One must now consider implementing hot aisle and cold aisle design, ceiling plenums, racks with built-in cooling and supplemental cooling.

Blades are here to stay, now the issue is how to control the ever increasing operating costs. Aside from data center design, it is equally necessary increase the utilization of resources already in place, and that is where virtualization provides a multi-dimensional solution.

Application intelligence can be distributed across multiple network-based platforms by application-layer routing protocols.

And, in principle, these systems are simpler from an operational perspective because the use of dynamic routing protocols at the application layer provides autoprovisioning of new users and services without extensive manual configuration.

At the most basic level, application-aware networking refers to the intelligent network-based processing of application-layer content – for example, doing message routing based on content, such as by reading a food-service order to route it to the correct airport terminal.

Application-aware networking products are not the same as Layers 4 to 7 switches, or content networking products. Such products typically switch on specific fields in the HTTP header (for example, the URI ), but they don't look into the application-layer payload itself.

But now Cisco Systems, Juniper Networks, Microsoft, and a host of startups are meddling with nature by selling programmable platforms that can understand and process the data running over them, moving smarts back from servers and other endpoints.

Cisco has set its sights first on shattering servers into pools of resources, like CPUs and memory, that are networked. It envisages a future in which the enterprise data center is reduced to little more than a virtual machine running on a giant switch--which, of course, it will provide. Juniper is giving its switches an API that third-party applications can access. 3Com is partnering with VMware to put virtual servers inside routers, and startups are scheming to steal the smarts from a huge variety of endpoints, from radio frequency identification readers to wireless sensors. Even IBM is joining the trend, selling appliances that off-load XML processing from servers and accelerate its WebSphere middleware using dedicated silicon.

While a single application or appliance may not need enough bandwidth to justify a switch slot, multiple applications or virtual appliances aggregated together are another matter. While a traditional architecture might use a 16-port Gigabit Ethernet module to connect one switch to 16 servers, the virtual data center would put 16 virtual machines right on the module. These VMs will have little relation to what we now think of as servers. For maximum flexibility, workloads must be able to move among servers--or switches--without necessarily dragging data with them, so they won't contain hard disks, relying instead on storage area networks. Cisco thinks the same will happen to memory: It will migrate outside the server or VM, requiring a new type of network to link memory and CPU

Next-Gen Data Centers

Generally speaking, application-aware storage is a policy manager implemented as a software layer installed on top of storage, and on top of storage management utilities. The policy manager extracts the established technical rules and business policies necessary to make rudimentary decisions. "A policy manager is no good unless it's got good, viable rules and information to act upon those rules," Schulz says. "This allows the policy manager [application-aware storage software] to act upon those rules."

This report is simply a must have for those wishing to reduce the overall costs of their Data Center.

IMEX Research, 1474 Camino Robles San Jose, CA 95120 (408) 268-0800 http://www.imexresearch.com
If you wish not to receive any IT information/news from IMEX, please click here