Comments for Lydia Leong

Comment on Rackspace Cloud Files (Mosso) + Limelight CDN by Mike

Mike — Fri, 20 Nov 2009 16:36:02 +0000

Lydia-

Thank you for posting this information. We have been evaluating a short list of hosting providers and content delivery networks and are leaning towards the Rackspace Limelight solution. This post confirms our internal opinion about choosing Rackspace to host our site in addition to hundreds of our customers’ sites. We will definitely benefit from your advice to use cnames as well. Thanks!

Comment on Job-based vs. request-based computing by Tyler012

Tyler012 — Sat, 14 Nov 2009 06:17:21 +0000

Good post. The information given in the post is so good and useful for the people to know more about the Computing Jobs and services. Thank you for the article given above…

Comment on TCO tool for cloud computing by Michael Blythe

Michael Blythe — Mon, 09 Nov 2009 11:24:35 +0000

There appears to be a lack of research relating to TCO of cloud computing services and the ‘real’ cost of moving from a traditional model to the hosted approach. I will be developing a number of TCO based models based on my experience so feel free to contact me for more information if you wish.

Regards,

Michael Blythe

Comment on Traffic Server returns from the dead by Traffic Server returns from the dead « Blogging

Traffic Server returns from the dead « Blogging — Tue, 03 Nov 2009 16:08:50 +0000

[...] Here is the original post: Traffic Server returns from the dead [...]

Comment on The Magic Quadrant, Amazon, and confusion by Managed hosting

Managed hosting — Fri, 30 Oct 2009 05:43:11 +0000

There are many quality people at Gartner but I’m afraid to admit that you are right on this one. The world of more informed users really highlights any kind of head fake that the big firms do to pay the bills.

Comment on Are multiple cloud APIs bad? by Do we need Cloud API Standards ? « The "Present" I live in

Do we need Cloud API Standards ? « The "Present" I live in — Sat, 24 Oct 2009 11:24:22 +0000

[...] “Are Multiple Cloud APIs Bad ?” – Lydia Leong (from Gartner) wonders if the standard should be around already popular Amazon’s S3 and EC2 API. Also brings about an important point (which I would talk about further below) that Rackspace API is different from Amazon because Rackspace has taken some fundamentally different approacheses. [...]

Comment on Jim Cramer’s “Death of the Data Center” by Chuck Hollis

Chuck Hollis — Fri, 23 Oct 2009 19:13:00 +0000

As I deal with the investment community frequently, this sort of thinking can be pervasive. Wrong, but pervasive

I wrote a post on this a while back — “The Paradox of Positive Elasticity” just because I found myself answering the same question repeatedly.

You might find it useful?

http://chucksblog.emc.com/chucks_blog/2009/10/the-paradox-of-positive-elasticity.html

Comment on The Magic Quadrant, Amazon, and confusion by Magic Quadrant or Magic Numbers? A Judge Will Decide. « The Effective Marketer

Wed, 21 Oct 2009 17:31:58 +0000

[...] to name a few). Gartner analysts try as they might to correct what they see as misunderstandings again and again are facing an uphill [...]

Comment on Cloudy inquiry trends by Lydia Leong

Lydia Leong — Tue, 01 Sep 2009 02:55:06 +0000

Broadly, I agree. Gartner’s position is that cloud is the IT delivery model of the future, and although a hybrid model will (and should) prevail for many years, the future lies in what we term tera-architectures — true computing as a utility. It will be a long journey to get there, but the fundamental value proposition is sound.

Comment on Cloudy inquiry trends by Dr. William L. Bain -- Founder and CEO, ScaleOut Software

Dr. William L. Bain -- Founder and CEO, ScaleOut Software — Mon, 31 Aug 2009 16:36:41 +0000

Lydia, although the concept of cloud computing does carry with it inflated expectations today, it’s important to note that it’s a nascent technology which over time should become an integral and essential platform for hosting a wide variety of mission-critical business and scientific applications. If you look past the hype, cloud computing at its core can offer beneficial resources that are just too attractive to undervalue, such as offloading data center management and providing access to an elastic pool of computing. Moreover, the ability to virtualize these resources and consolidate them into a centrally managed data center offers enormous economies of scale and energy savings.

We do agree with you that there are several key challenges today that must be overcome if cloud computing is to deliver on these expectations. For example, applications which run in data centers on-premises must be able to seamlessly and securely migrate into the cloud without the need to redesign their data storage architectures. Today this is difficult to do because security, deployment, and storage mechanisms are unique to each cloud vendor’s infrastructure and distinct from mechanisms typically used on-premises. Once hosted in the cloud, applications also must be able to achieve their performance and reliability goals while running in a virtualized environment. Today’s virtualized infrastructure needs to evolve so that it can precisely allocate the virtualized resources (CPU, memory, storage, and networking) needed by applications need to meet performance goals.

With regard to storage, today there is no way to seamlessly migrate data into the cloud as an application scales from an on-premises deployment to an elastic set of virtual servers running within the cloud. Data must be packaged and staged in a cloud-based “blob” store or database server for retrieval within the cloud. For example, an active, e-commerce Web site cannot scale into the cloud and reach across to retrieve shopping carts that originally are hosted on-premises. ScaleOut Software is attacking this problem with distributed data grids than span multiple sites (e.g., on-premises and cloud) and transparently migrate application data where needed at all times.

Once these early obstacles are overcome, we believe that cloud computing will offer an enormous opportunity to realize its full potential for efficient, scalable computing. Its unique ability to provide an elastic pool of computing resources will enable applications designed for scalability to automatically harness additional resources as needed to handle very large workloads. These applications will be able to effortlessly process very large data sets in parallel, which we expect to have a revolutionary impact on data analysis. As they say, the best is yet to come, and I would expect that the inquiries you are receiving will only increase as the benefits associated with the cloud materialize.