September 2012 Archives

Paradigm 13 Shift 2 SDN (lost count)

By Steve Broadhead on September 28, 2012 8:43 AM | No Comments | No TrackBacks
| More
Such was the count at the end of Day 1 of Netevents Portugal. Thirteen "paradigm's" and two "paradigm shifts". Surprisingly there were no "out of the boxes" and only one "granularity" reference. It should also be noted that the "p" word" was used by at least four different nationalities, so it's not a single country syndrome.

But the winner for "fully embraced buzz-phrase" has to be SDN or Software Defined Network, something we've spoken about in this blog on more than one occasion, including yesterday. The thing is, regardless of whether it is simply what network management should have been all along (or not), something really IS going on here - real products, services, open (OpenFlow) and proprietary (the rest) and pilot customers. It is, therefore, realistic to suggest that we are going to move into phase III of IT; mainframe, then networked and now SDN - i.e. fully separating the control and management of network traffic (users and application) from the physical components - switches, routers etc. For it to be truly worthwhile, SDN has to enable us to manage the network on a per user, per application, per connection (end to end) basis.

Is this feasible? Certainly. Does anyone have a true, fully-working solution? Watch this space - at least until Wednesday, when I can reveal one client of Broadband-Testing who has all the components; now just let us loose on the testing of said solution...

IT Is Prawn Cocktail ?

By Steve Broadhead on September 27, 2012 10:41 AM | No Comments | No TrackBacks
| More
Bem Vindo from the Algarve, at the latest Netevents symposium.

One of my favourite topics in networking (and IT in general) is how often we revisit old "recipes". In the same way that prawn cocktail has become trendy again, so it is with networking and Netevents. Two panel debates in, we've already had seven "paradigms" (IT buzzword of the year, 1995) and several "visions" and a few "hype cycles".

Debate topics are pretty well predictable:

- BYOD
- SDN
-  Mobile + Cloud = opportunity or risk?
etc...

The focus of the BYOD debate (and let's face it, people have been bringing their personal laptop into work and copying data onto it to work on from home out of office hours since the early '90's) was device management and security. But is the real issue here not the device, but the kind of applications that people are using on them, and adopting and managing those?   In other words, at what point do applications such as Facebook become "enterprise" applications and how do we then manage those, rather than simply block them (and the devices themselves)?

Now we're onto the subject of SDN - Software Defined Networking. The panel talk is about automation, removing the need for manual administration, control of mixed vendor networks etc. Isn't this called vendor-independent Network Management - i.e. what all the net' management vendors in the early '90's set out to achieve? So, it didn't get there - will SDN?

The debate goes on...



 

Tech Trailblazers Update

By Steve Broadhead on September 4, 2012 3:15 PM | No Comments | No TrackBacks
| More
Just a quickie update to all you vendors with mega technology out there re: the Tech Trailblazer awards wot I blogged about earlier this summer.

Entry levels have proved (as did Top Gear) that you can't have too many awards competitions, and these are still open until the 12th September (for late birds, the early one has closed) in the following categories, just to remind you all:

  • Big Data Trailblazers
  • Cloud Trailblazers
  • Emerging Markets Trailblazers
  • Mobile Technology Trailblazers
  • Networking Trailblazers
  • Security Trailblazers
  • Storage Trailblazers
  • Sustainable IT Trailblazers
  • Virtualization Trailblazers
There's over a million dollars up for grabs, so well worth the entry. To do so, just go to:


Simple as...

Enhanced by Zemanta

At The End Of The Network

By Steve Broadhead on September 4, 2012 3:09 PM | No Comments | No TrackBacks
| More

One of the problems we've faced in trying to maximise throughput in the past has not been at the network - say WAN - level, but what happens once you get that (big) data off the network and try to store at the same speed directly onto the storage.

 

We saw this limitation, for example, last year, when testing with Isilon and Talon Data and using traditional storage technology - the 10gigabit line speeds we were achieving with the Talon Data just couldn't be sustained when transferring all that data onto the storage cluster. While we believe that regular SSD (Solid State Disk) technology would have provided a slight improvement, we still wouldn't have been talking end-to-end consistent, top-level performance.

 

So it's with some interest - to say the least - that I've started working with a US start-up, Constant Velocity Technology, that reckons it has the capability to solve exactly this problem. We're currently looking to put together a test with them:  http://johnpaulmatlick.wix.com/cvt-web-site-iii - and another "big data" high-speed transfer technology client of mine, Bitspeed, with a view to proving we can do 10Gbps, end-to-end, from disk to disk.

 

Even more interesting, this is happening in "Hollywood" in one of the big-name M&E companies there. However, if any of you reading this are server vendors, then please get in touch as we need a pair of serious servers (without storage) to assist with the project!

 

Life beyond networking...

Enhanced by Zemanta
« July 2012 | Main Index | Archives | October 2012 »