Archive for the 'Cloud computing' Category

Cloud, How Big Is It?

Cloud QuestionCLOUD, HOW BIG IS IT?

The Cloud, it’s Huge! But what does that mean? Let’s start with numbers we are familiar with, then see how it grows to numbers we have a harder time getting our head around:

  • Starting with a binary digit, or “bit”
  • 8 bits makes a “byte” or a character on a page
  • Then about a thousand of those, or a “kilobyte”. Two kilobytes is about a page of information
  • Then add 3 zeros to the end of that, or a “megabyte.” Two or three megabytes is about the size of a digital song
  • Adding another 3 zeros to the end of that and you have a “gigabyte.” Computer laptop memory (RAM) and iPod capacities are usually measured in Gigs
  • 3 more zeros on the end of that and you have a “terabyte.” Computer hard drives (disks) are measured in TBs
  • Another 3 more zeros and you have a “petabyte.” The capacity of large data storage arrays are measured in PBs.
  • When you add 3 more zeros or orders of magnitude (powers of ten) you have an “exabyte”. One of those represents the amount of mobile data traffic used in the US last year, according to an analyst
  • Another 3 zeros gets us to the “zettabyte” scale, a number not yet in common parlance, but that represents about the entire global digital data used back in 2009
  • 35 of those “zettabytes” is the forecast for digital data use by 2020, just 9 years from now
  • About a third of that will be in the Cloud.

A “yottabyte”? No relation to the Jedi Master.

Click on the image below to see it at full size.

Cloud Math

Thanks for coming along.

BillPetro.com

Cloud OS: VMware vSphere 4 Launches

CLOUD OS: VSPHERE 4 LAUNCHES

This week, VMware upgraded its main product line VMware Infrastructure 3, to deliver IT infrastructure as a service internally.

SaaS, or “Software as a Service” is a term often seen in discussions of Cloud Computing, where these services are offered remotely from data centers over the Web. With vSphere, companies can now do this internally as well. VMware is calling it “the industry’s first operating system for building the internal cloud.”

vSphere was foreshadowed with the announcement of  VDC-OS (Virtual Data Center Operating System) last Summer at the Las Vegas VMworld Conference, as I wrote about here.

Virtualization is a disruptive technology of sorts, in that it breaks the “hard link” between hardware and software, or more specifically between the Operating System and hardware. A VMware virtual machine (VM) can create an OS that is essentially hardware vendor independent — as long as it runs on x86 processors — and encapsulate it to run on any number of platforms. Indeed multiple OS instantiations can run on a single physical server. The impact of this is phenomenal: it can undermine the value proposition of an “integrated” solution from some server vendors. In this way, it can reduce capital and operating expenditures, even as information explodes.

There are several capabilities of this new release that are impressive. Here are the ones that caught my eye:

Efficiency:

  • 8000 DB transactions per second per virtual machine. Can you say OLTP?
  • Over 200,000 IOPS per ESX host. Can you say Ultra Enterprise Sun IBM “Oracle”-Fire 15000 High Performance Computing? [I launched the StarFire 10000 when I was at Sun.]

Control:

  • Ability to control an entire virtual data center from a single pane of glass. VMware calls this the ability to “holistically manage” infrastructure elements.
  • Access to the Cisco Nexus 1000v “virtual switch.”

Choice:

  • Storage VMotion permits moving the VM from where it lives — on the storage — to other kinds of storage as needs require. This would allow migrating for example, from expensive Fibre Channel disk drives to less expensive SAS drives, or even changing storage protocols.
  • Fault Tolerance permits a “ghost” version of a VM to run on another physical server, in “lock step” but invisibly, such that should the original VM or server fail, the fault tolerant version will immediately become available on the second server.


Thanks for coming along.

BillPetro.com

Cloud Optimized Storage: Atmos

In 1969 during the original Star Trek TV show, The Cloud Minders episode featured Stratos, a city in the clouds. Today, EMC launches Atmos, cloud optimized storage. This is not science fiction, but the realization of much work on managing very large amounts of data “in the cloud.” Having been in the hands of customers since early Summer, this offering is intended to manage massive amounts of Internet-based information. How massive? Petabytes to begin with.

Forecast: Cloudy?

“Cloud computing” is a very popular topic right now, and there is often more heat than light shed on the subject. EMC releases Atmos at a time when many others are still debating over cloud computing and arguing whether it is or is not grid computing, software as a service, outsourcing, or Web 2.0-based data.

The Atmos sphere

What does Atmos encompass? First, it is different than simply block-based Storage Area Networks (SAN), file-based Network Accessed Storage (NAS), or even object-based Content Addressed Storage (CAS). EMC introduced CAS in the Spring of 2002 with the launch of Centera, part of an object-based archiving and compliance solution. As revolutionary as this was half a decade ago, so too is COS today.

  • Massive Scalability

Global manageability through a namespace, regardless of geographical location. Multi-tenancy means that different tenants could store their unique information objects in their own private namespace under the Atmos namespace.

  • Policy Based Information Management

Information can be managed by policy, and automatically acted upon by business rules for the metadata that defines the policy. For example, if an object conforms to a policy of “popular”, it could be replicated appropriately to deliver better service levels to users. When demand drops off and it becomes less “popular” the number of copies could be automatically pruned.

  • Operational Efficiency

Whether it’s a drive, server, or network — all of this can be seen and managed from a single console.

 

There are many excellent articles available already that explain this in greater detail. Check out Chuck Hollis at Chuck’s Blog, Mark Twomey at Storagezilla, Steve Todd at Information Playground, Dave Spencer at Dave Talks Shop, Dave Graham at Dave Graham’s Weblog, and Len Devanna’s summary.

Thanks for coming along.

BillPetro.com

VMworld 2008: Day 1 Keynote

Paul MaritzVMWORLD 2008: DAY 1 KEYNOTE

Following yesterday’s opening event, both the Technology Exchange and Partner Day, the conference started in earnest today with a keynote by Paul Maritz, VMware President and CEO.

Paul is quite an articulate speaker, sounding both like a savvy businessman and an erudite professor. For many, his accent is difficult to place, is it Australian, South African? Turns out he was born in Zimbabwe, next door to South Africa, but he went to school in Cape Town, South Africa. Someone remarked that he pronounces some words like Sean Connery. Paul did work at University of St. Andrews in Scotland. Sean Connery grew up in nearby Edinburgh, Scotland.

Back to the keynote, Paul amplified many of the concepts introduced and announced yesterday, specifically by drawing on the past to explain where VMware is going in the future. I found the history a good review of many of the events I’d witnessed in my own long career with computing, and a fascinating basis for describing where he sees the company going, as well as a foundation for the many announcements this week. How did he do this?

Two Models

He went back to the ’60s and ’70s (mainframes) and outlined the contrast between the Centralized vs. De-centralized models of computing. While he did not say so, the industry has swung between them several times over the last 30 years. He pointed out that we initially saw mainframes in the early days, and the proliferation of PCs in the ’80s. The early ’90s saw the rise of x86 Servers as well as the rise of the Client/Server model.

A Third Alternative

He quipped that it’s ironic that he is now profiting from his previous sins in promoting the Client/Server model, which got us into a world of hurt so that we now seek another path, the best of both the Centralized and the De-centralized models. The advent of the Web in the mid-’90s offered the promise of this. He paid tribute to the founders of VMware, who started the company ten years ago back in 1998.

VMware’s Initial Offerings

He described VMware’s early efforts with both VMware Workstation (on the client, or De-centralized side in 1999) as well as the early VMware Server (GSX, on the Centralized side in 2000.) He pointed out that many other companies are currently virtualizing on the Centralized (server) side, but reminded us that VMware has done both sides, but raised the bar in 2004 with the introduction of VMware Infrastructure, a higher level of abstraction than either Server or Client side virtualization.

Paul pointed out that the “Best of Both” on the Web is now more promising with a variety of new (Web 2.0) technologies like AJAX (Asynchronous JavaScript And XML), Ruby on Rails, and Python. 2007 saw the launch of VMware’s Fusion, virtualization of Windows on the Macintosh and the popularity of Cloud computing.

It’s The Platform

At this point, I reflected that during my career I’d seen software engineers write to whatever the leading “platform” was at the time. In the ’80s it was Unix, and particularly SunOS (Solaris). The virtuous cycle had the platform supporting applications that led to more volume… which made the platform more viable. With the rise in popularity of Windows, it became the platform of choice. In the mid-’90s it was Java, with the promise of “Write Once, Run Anywhere” across a variety of devices. But last year, it became clear to me, especially following VMworld 2007 in San Francisco, that VMware was becoming the “platform”. Both the Press and the Analysts “got it,” and Wall Street saw a huge jump in VMware’s stock price following the ESXi announcement and others.

The New Platform

VDC-OS So, what’s the new “platform”? Paul explained the Virtual Datacenter OS, or VDC-OS. It is a way to support a variety of current popular “platforms” line .Net, Windows, Linux, Java, Software As A Service… with Application vServices that provide Availability, Security, and Scalability. This rests upon Infrastructure vServices called vCompute, vStorage, and vNetwork… as well as Cloud vServices. These in turn live on either the On-premise cloud, or an Off-premise Cloud (for additional resources.) Meanwhile, all of this can be managed by vCenter (the rebranded Virtual Center management framework) which handles both Application Management at the top end and Infrastructure Management at the backend.

To this end, many announcements fit into this new, higher level of abstraction. Once again, VMware raises the bar.

Thanks for coming along,

BillPetro.com

© Bill Petro – visit the author for more great content.

EMC World: Day 2 Recap

lasvegas_sign.jpgEMC WORLD: DAY 2 RECAP

EMC World, in its 2nd day in Las Vegas, is by and for engineers. While there are more suits present than in previous years, nevertheless most of the presenters are usually not polished marketing presenters but often the software developer who wrote the code for the product. This provides a level of unparalleled access for attendees.

I had lunch with a SAN Administrator for a state government office and we discussed data replication at a pretty deep level. This gentleman was a heavy user of the technology and was particularly impressed with EMC’s RecoverPoint technology. I asked if he found the price-point to be prohibitive. He assured me that it was quite attractive: the price had dropped since the product was initially released, and it offered so many features and such great capabilities that it eliminated the need for spending money on lots of other technologies. He found it to be ideal in a VMware environment and was excellent in the event he needed disaster recovery.

I’m typing this article in the Cyber Cafe, a number of stands with numerous laptops set up for attendees to use between talks and strolls through the Solutions Pavilion (Exhibit Hall). The EMC sysadmins have done a great job setting them up, with easy access to web browsers. They told me, however, that some will unplug them so that they can plug in their own laptops to charge them. Bad idea. The batteries have been removed from these laptops — I’m told they tend to grow legs and walk away — and as soon as they are unplugged, they shut down.

Speaking of the Pavilion, I spent some time speaking to a variety of exhibitors today. Not only are there “the usual suspects” but as in years past, some competitors on the floor as well, showing the openness of the show. Here’s what caught my eye and ear:

  • ss4200e.jpgIntel SS4200-E

With a name that rolls off the tongue, this is the Intel version of the product I mentioned in my previous post yesterday that Dave Donatelli referenced as the perfect graduation present — a sub-$500 external Network Attached Storage (NAS) device for consumers, that leverages EMC’s LifeLine software. The demo is a device that takes 1-4 internal drives (sold separately) that can provide RAID protection as well as a variety of media services. For example the device can speak to an Xbox as a bridge, display movies to a widescreen TV… display photos to a digital picture frame… provide web-accessible photos to a remote device like an iPhone… play MP3s through an iTunes server… etc.

  • iomegacenterlogo_ec65.pngIomega

While slightly later to market, and with the intention of addressing a slightly different market than Intel, Iomega — currently being acquired by EMC — also previewed their LifeLine-based offering. It will be available in different configurations than the Intel offering, but with all the same capabilities. This is one to keep an eye out for.

  • retrospect.gifRetrospect

Acquired years ago by EMC as part of the Dantz acquisition, this well respected consumer and small business backup solution was, 20 years ago, the only 3rd party backup product for the Macintosh. This is features as part of LifeLine, and new features and capabilities are coming that will make this a compelling offering for both local and cloud-based backup

  • mozy_logo.pngMozy

Speaking of cloud-based backup, this popular technology, acquired last year by EMC, allows consumers and small businesses to have SAAS (Software As A Service) backup for pennies a day. An application on a Mac or PC backups files over the Internet in encrypted form, initially as a full backup, then block-level incrementals thereafter. Offsite backup without having to move backup media offsite.

  • 3D or Data De-Duplication

Mark Twomey, also known as Storagezilla, was in the EMC booth explaining the new data de-duplication capabilities of the EMC Disk Libraries. Mark knows everyone.

I’ve reported more details via Twitter. These can be found either at:

Thanks for coming along.

BillPetro.com

© Bill Petro – visit the author for more great content.