How Can a Software-First Strategy Improve Your Data Center?

Software-first networking can be an interesting theory or a transformative technology for enterprise computing. Which one it is depends on your perspective…and maybe your age.

Many of today’s enterprise IT leaders will remember the glory days of infrastructure expansion. If you needed more CPU or storage, it was easy and relatively inexpensive to throw new hardware at the problem of scale and capacity.

While that approach was not sustainable, it helped to establish a habitual response to IT challenges: more hardware. For a while it seemed that any computing, storage, or networking problem could be solved by more or better boxes.

But the modern IT landscape has blown that assumption out of the water. Data volumes and encryption demands are orders of magnitude larger than they were a mere two decades ago. Computing power has increased exponentially and predictably such that dual-socket servers with 44 cores (88 threads) are now readily available.  Eye-popping volumes of corporate data and off-the-charts processing power can be a powerful combination…if your network can move data efficiently enough.

Enter the software-first approach. It is a strategy that removes barriers that arise when IT services are bound to physical devices. While you can’t entirely ignore hardware, the right components can prepare you for the future of software-defined everything. Intel and Citrix are working together to help you make a successful transition from a hardware-centric network into a software-defined network (SDN).

Hardware Versus Software

A hardware-defined network has inherent limitations associated with physical devices. These limitations constrict data flow and make  it harder to scale, upgrade, or otherwise transform your data center.

A software-first strategy shifts your company’s infrastructure from being overly reliant on hardware. The strategy looks to software solutions, rather than more hardware, to give your data center the agility and throughput that it needs to process today’s gigantic volumes of structured and unstructured data—even when that data is encrypted.

One way to begin the software-first journey is to incorporate automation into the network by way of application delivery controllers (ADCs).

You might be thinking, “But, Tim, you just told me that moving toward a software-first strategy is the way to go. An ADC is a piece of hardware.” You’re right—at least partially. You can’t replace your hardware-centric data center with a software-centric one over night, but you can take small steps toward your software-defined infrastructure (SDI) goal.

Your First Steps

If you want to move toward an SDN, you’re going to want Citrix NetScaler. NetScaler is an ADC that can be deployed as a physical, virtual, physical-virtual hybrid, or containerized appliance. Because each iteration of NetScaler shares a single API and set of code, it’s much easier to transition from a physical NetScaler ADC to a virtual ADC than it would be to install or upgrade hardware. The shared code base and API also mean that NetScaler integrates with your heterogeneous environment regardless of hypervisor, cloud, orchestration platform, or fabric architecture.

For example, you could deploy Citrix NetScaler MPX today—that’s the bare-metal version—to optimize network traffic. Then when you’re ready to start walking the software-defined road, you can switch to Citrix NetScaler SDX, which is a physical appliance that supports multiple virtual instances of NetScaler software. Then, once you’re ready to move your network control into the virtual space, you can upgrade to Citrix NetScaler VPX, a virtual appliance.

If you’re a cloud-based provider or app developer, the micro-version of NetScaler VPX, known as NetScaler CPX, might be your best bet.

Software-first Support

Running NetScaler on the Intel Xeon processor E5 family can propel you into the software-first arena. NetScaler was designed to capitalize on technology built into the Intel Xeon processor E5 v4 family allowing NetScaler to bypass main memory and route data through to the L3 cache, which can reach sizes up to 55 MB. That puts data even closer to the processor.

The Intel Xeon processor E5 family is available in a variety of configurations, so you can choose the processor that best fits your data center’s needs. The Intel Xeon processor E5 v4 family supports up to 1.5 TB of fast memory (DDR3/DDR4). Other Intel technologies baked into the processor enhance security and accelerate encryption.

Read More
There’s a lot more to the symbiosis between the Intel Xeon processor E5 family and Citrix NetScaler than this blog can cover. Read our latest paper, Citrix and Intel Help Advance Next-Generation Data Centers with NetScaler, to get the full picture.

To stay up to date on the latest big data news and developments, check out the #TechTim community on Twitter, and follow me @TimIntel.

Published on Categories Cloud Computing, Data Center, Software Defined InfrastructureTags , ,
Tim Allen

About Tim Allen

Tim is a strategic marketing manager for Intel with specific responsibilities related to the cloud, big data, analytics, datacenter appliances and RISC migration. Tim has 20+ years of industry experience including work as a systems analyst, developer, system adminstrator, enterprise systems trainer, product marketing engineer and marketing program manager. Prior to Intel Tim worked at Tektronix, IBM, Intersolv, Sequent and Con-Way Logistics. Tim holds a BSEE in computer engineering from BYU, PMP certification and a MBA in finance from the University of Portland. Specialties include - PMP, MCSE, CNA, HP-UX, AIX, Shell, Perl, C++