ZK Research: Home
Google+
Twitter
LinkedIn
Facebook
RSS Feed

Posts Tagged ‘data center’

This morning, Cisco announced its intention to acquire Whippany, New Jersey-based WHIPTAIL for $415M. For those who don’t know WHIPTAIL, the Cisco press release describes the company as a leader in “high-performance, scalable solid state memory systems.”

However, if you look at the WHIPTAIL website, the company describes its products as “The First Family of High Performance Storage.” So, which one is it? Well, both really, as the company offers highly scalable, flash-based storage arrays, and if you look at the specs on these products, they’re fast. How fast? Well, WHIPTAIL can move data faster than the Buffalo Bills can choke away a fourth-quarter lead (I included this just for Duffy). I think Cisco is being careful in its choice of words, but I’ll get to that in a bit.

First, the WHIPTAIL products – the company has three main products to its storage family. The high-end system can scale up to 360TB of storage at 4 million IOPS. The low end of the product line is a product that ranges from 3-12 TB at 250K IOPS, so it has quite the range of from the low to high end. These systems aren’t designed to replace massive enterprise storage systems but rather to be used where speed of storage is critical. Bare metal applications, video transcoding and analytics, such as HANA, come to mind. The continued growth of digital content combined with the Internet of Things will increase demand for big data and analytics, requiring faster-performing data access, which is the value proposition of WHIPTAIL.

[keep reading…]

The data center is where all the action has been in networking over the past few years. We saw the introduction of the network fabric, the rise of software defined networks (SDN), a number of startups emerge, and we’ve seen a fair bit of M&A activity as well. Because of the rapid evolution, we’ve seen almost every major network vendor – Cisco, Brocade, Juniper, Extreme, Avaya, Alcatel-Lucent and others – revamp the data center portfolio.

The one vendor that I thought was noticeably absent from the data center networking wars was HP. The company outlined its FlexFabric vision last year, but the only products it had to support the related architecture was the 10K, which is a campus switch, and the 12,500, which was great when H3C first released it, but was getting a bit old even when HP acquired H3C. Now, it’s clearly past its prime. The company has positioned the 12,500 as a data center switch and has beefed up the features set accordingly. The 12,500 now supports Ethernet Virtual Interface, SPB and other data center features. As of now, it’s limited to 10 Gig-E, but HP has stated that 40/100 Gig-E will be available later this year. Despite the added features, though, HP is the vendor I get the least amount of inquiry on regarding data center networking.

[keep reading…]

This week, data center specialist Brocade announced its “HyperEdge” architecture for campus networks. The concept of HyperEdge is similar to the value proposition the company put forth with its data center fabric architecture – networking is becoming increasingly complex, so a simpler, flatter network is required to support companies moving forward.

Over the past few years, the concept of the network fabric has been aligned with the data center since this is where the most significant changes have been on the compute side. Virtualization, cloud computing, growth in storage and other trends have driven more East-West traffic, creating the need to move away from the traditional multi-tier, spanning tree (STP) supported network. The solutions offered by almost every mainstream network vendor today is to implement a two-tier network (or single-tier in the case of Juniper’s QFabric) based on TRILL, shortest path bridging or some sort of proprietary protocol to replace STP.

[keep reading…]

Last week, the esteemed Jim Duffy wrote this news article covering the release of the Juniper EX9200 data center switch. The article left many questions in my mind, including when should customers buy an 8200 now? What about QFabric? What types of customers are right for the beefiness of a product like the 9200? I had a chance to talk to Juniper about the product and I thought I’d clear up some questions that were still outstanding.

The first notable point with the EX 9200 is that it’s actually based on the MX router, not the EX 8200 switch line, meaning it’s an EX by name only. This seemed a bit odd at first until you understand that the primary use case for the 9200 is as an SDN switch, making programmability a must-have. The EX line was built on merchant silicon, whereas the MX was built on the highly programmable, custom Trio ASIC, giving Juniper the foundation it needs to build a programmable switch. With this product, Juniper hopes to one up its Tasman Ave-based friends over at Cisco. Cisco’s programmability story is centered on the python-based OnePK programming environment. Juniper’s Trio-based 9200 gives the company the ability to program at a chip level, allowing for custom rewrite of the silicon. This should give Juniper some unique capabilities. However, it only does so for one of its three switching lines, whereas Cisco’s OnePK covers many of its products today, including the upcoming new ISRs, allowing Cisco to continue to promote the concept of an “architecture.”

[keep reading…]

Most of the focus of software-defined networks (SDNs) has been on how it impacts the layer 2/3 switch vendors. The industry seems to have moved off of this notion that it commoditizes the underlying infrastructure, but recently another question has come up. Big Switch recently launched the company and related products, one of which is called “Big Tap,” that provide traffic visibility functionality similar to what one might get from vendors such as Gigamon and VSS. This has raised a question: are SDNs a death knell to the traffic visibility vendors?

I looked at this and then talked to a number of customers, including Big Switch, and I believe the information that one can get out of an SDN-led product to be very much complementary to the traffic visibility market, not competitive. Think of “Big Tap” as being traffic visibility light where they provide a very basic level of information. The level of information that one gets from the dedicated vendors is much richer and more granular than what one would get from Big Tap.

[keep reading…]

ZK Research is proudly powered by WordPress | Entries (RSS) | Comments (RSS) | Custom Theme by The Website Taylor