clouding computing


Right after hemorrhaging dollars for the greater portion of a decade and submitting for Chapter eleven bankruptcy twice in three years, it seems like storied server and supercomputer maker SGI could really turn a revenue yet again. In April 2009, concurrently with its second Chapter eleven submitting, SGI sold by itself for peanuts ($ 25  million) to webscale server maker Rackable Techniques, which — surprisingly — decided not only to keep promoting SGI’s merchandise but in fact to get on the fairly-tarnished identify of the then-NASDAQ-delisted entity. Now, right after an to begin with bumpy ride that seems to be like the outcome of acquiring to iron out some wrinkles and soak up SGI’s higher-conclude products lines, the new SGI noted remarkable profits during the 2nd quarter and elevated its fiscal 12 months 2011 guidance from “break-even” to “profitable.”

The methods the new SGI used to flip a profit from the aged SGI’s cash-dropping company mirror some simple guidelines that your stock adviser, or your father, may possibly have advised you about how to make money with your investments. You want to buy inexpensive, put some income in blue chips and diversify. Of course, acquiring a significant, unprofitable business is more than just a financial investment — it’s also a enterprise — and that signifies finding operational fees in examine by any signifies needed.

Get inexpensive

Rackable absolutely bought SGI inexpensive. The $ 25 million cost tag (technically greater due to the fact Rackable assumed specific SGI liabilities that weren’t discharged) is a mere fraction of the hundreds of hundreds of thousands in revenue that the old SGI used to bring in quarterly. It’s noteworthy, nonetheless, that the new SGI nonetheless isn’t reaping old-stage-revenues from its meager investment — but its $ 177.five million in 2nd-quarter earnings signify an 88 % 12 months-more than-12 months boost and a file quarter for the new SGI.

There’s a lesson to be learned here, which seems to be the virtues of patience. Technological innovation organizations this sort of as SGI — or even Sun Microsystems, which Oracle arguably acquired for a steal at $ 7.4 billion — didn’t achieve the peaks they did with no wonderful technology and even larger minds. But when their innovation-at-all-costs business types and poor decisions start off dragging them down financially, there are rewards to be reaped for vendors ready to set forth the hard work to quit the bleeding. Not every single one particular is headed for bankruptcy — which surely eases a whole lot of the financial debt soreness and takes shareholders out of the picture — but when they’re at their lowest, even shareholders may possibly accept a lower supply to get one thing from their investments.

Don’t low cost blue chips

Higher-performance computing is considerably of a blue-chip market in the technological innovation entire world, as there is a stable core of organizations and firms that will need depend on it and will often be inclined to shell out the large bucks essential to get tremendous-effective systems. So, instead than try out to just integrate SGI’s technologies into the present Rackable organization of promoting webscale servers, the new SGI embraced HPC and is as soon as once more a serious player in that space. According to CEO Mark Barrenechea, technical computing (SGI’s favored expression for the space, which Barrenechea says encompasses HPC and a total good deal a lot more) is a $ nine billion industry with Dell operating at the reduced conclude, IBM running at the quite high end with its proprietary Power7-primarily based systems, and with the center extensive open for the taking by SGI and its lineup of Altix methods and substantial-overall performance storage gear.

In truth, Barrenechea advised me that about 50 % of SGI’s organization comes still arrives from the public sector, like the Division of Vitality laboratories, intelligence companies and NSF-funded universities. “It’s a fantastic marketplace for us,” he noted, incorporating that SGI systems for these sorts of clients, as well as for significant enterprise consumers, typically pack among fifty and 100 teraflops per cabinet. SGI Altix methods accounted for 22 of the 500 fastest computer systems in the planet in accordance to November’s Top500 listing, and, heading ahead, SGI also expects to be “squarely in the center” of the race to exascale pcs. The Obama administration is proposing $ 126 million for exascale investigation in its congressional price range.

Diversify

The new SGI isn’t wholly dependent on selling huge shared-memory programs to national labs, although. In accordance to Barrenechea, the firm doesn’t break down earnings by divisions, but it expects only about one-3rd of yearly income to arrive from the traditional SGI merchandise, with the remaining two-thirds split equally amongst solutions and the Rackable business line, which focuses on dense, electricity-successful servers for webscale data centers. The business has by no means genuinely moved the needle in terms of overall server revenue, but it does have some higher-profile buyers — such as eBay, Carbonite, Microsoft, BT and Amazon — that contribute enough to the bottom line as they fill their info centers with Rackable gear. For the duration of a info-middle construct-out in 2008, Amazon purchased $ 86 million value of Rackable servers.

The unsightly facet

However, producing an unprofitable organization lucrative isn’t often quite. Much like Oracle axed many Sun goods and staff on taking above, on best of placing price tags on as soon as-free of charge merchandise this sort of as MySQL, the new SGI has had to trim some body fat, too. For a single, Barrenechea explained that the aged SGI spent a lot of funds on projects that weren’t often tied to any immediate earnings, whereas the new SGI tries to concentrate on consumer- and industry-focused projects that it expects will pay out spend off fiscally. A cynic may see this as killing accurate innovation in the name of creating a buck, a stance that has some merit.

And just this morning, information broke that SGI is laying off 55 employees, roughly4 % of its world wide workforce. I inquired no matter whether the layoffs are component SGI’s commitment to turn into lucrative in this fiscal yr, to which a firm spokesperson confirmed via e-mail that “[s]treamlining operations and prioritizing performance is immediately aligned with attaining SGI’s FY11 economic goals.”

Read more:

  • mark barrenechea is fat

Terracotta is bringing actual-time analytics to the masses (of Java customers, at minimum) by letting Ehcache users query info saved in the product’s in-memory cache. With Terracotta’s new Ehcache Search merchandise, buyers can perform simple queries in true time in opposition to as considerably as terabytes of data stored in their transactional caches,without having having to install any new servers or purchase new appliances. The strategy won’t substitute the data warehouse, but it could have a considerable impact on the potential of analytics software program growth.

The open-resource Ehcache is to Java what memcached is to dynamic world wide web languages, in that it lets builders retailer particular info in-memory to keep away from the inherent latency of interacting with the database every time an application requirements to serve a piece of info. This setup is fantastic for transactional workloads, but, normally, any analysis nonetheless calls for a journey to the database. For queries that could benefit from genuine-time benefits, this latency can turn into troublesome, particularly if the database is getting bombarded, and slowed down, by significant numbers of requests. Enter Ehcache Search.

According to Terracotta CEO Amit Pandey, 1 early consumer that manages logistics for a fast-meals chain was in a position to reduce latency periods to the sub-second assortment from almost a minute. Desperate for far better overall performance, the organization was taking into consideration Oracle’s higher-performance Exadata Database Machine, but didn’t require all the complexity, and didn’t genuinely want to pay out the large price or deal with the 12-month set up procedure, either. A computer software-only product, it took only a month to install Ehcache, load the sought after information into the memory of the existing software servers and start using the merchandise.

But, as even Pandey acknowledges, databases and information warehouses aren’t heading away. They’re still required for complicated queries, specially against huge volumes of data that just are not able to be stored in-memory. Despite the fact that, a Terracotta product sales rep may be fast to position out that the line is blurring. When used in mixture with Terracotta’s BigMemory product, customers can keep up to a terabyte of information in-memory (officially, despite the fact that Pandey says users are storing up to 4 TB), and the business is organizing to enrich the analytics capabilities in the following eighteen months. Presently, Ehcache Search is obtainable in each open-source and enterprise editions, and BigMemory is obtainable exclusively as an enterprise version.

This blend of transactional and analytical environments doesn’t start with Terracotta, however, and it won’t probably finish with it. Currently, SAP is marketing its High-Efficiency Analytics Appliance (HANA) that relies on in-memory processing to let consumers “instantly explore and review all of their transactional and analytical information,” and I have to think other vendors with their fingers in the two pots (e.g., Oracle and IBM will roll out their personal offerings, as properly. Pandey thinks they might even roll out light-weight versions in the very same vain as the open supply Ehcache Search, but stated that will require robust client desire. Contemplating those companies’ reliance on Java, and that Ehcache has a footprint hundreds of 1000's of Java applications, Terracotta might be the company that makes Oracle, IBM and SAP buyers see the light.

If that comes about, it could symbolize a real shift in the sophisticated analytics marketplace related to the freemium trend we’re presently seeing in the SaaS room. Presently, vendors such as Terracotta, EMC (by means of Greenplum), and Jaspersoft and Pentaho are all approaching no cost, open-resource analytics from various perspectives — real-time analytics, analytics database and BI, respectively — but acquiring massive software program vendors on board with giving away superior features to some diploma may possibly be regarded as a landscape shift that couldn’t have been imagined just many years ago.

Right after a $ fifteen.7 billion fourth quarter that put Dell about the $ 60 billion earnings mark for the yr, CEO Michael Dell got to speaking once once more about his firm’s acquisition tactic, suggesting much more computer software purchases are certainly on the way. According to a ZDNet post recapping Dell’s submit-earnings conference call, Dell stated the firm is targeting smaller acquisitions (read “not AMD”) and people that can add value to servers and storage gear, specially in bundle-sort bargains. This assertion, combined with current business happenings, reaffirms my belief that Dell will goal businesses comparable to these I suggested last month.

Just to be obvious, right here is Michael Dell’s quote in its entirety:

I think you are currently seeing us in methods management and some of the computer software all around supporting the programs infrastructure. We’re searching at how we can increase that platform into new consumer categories and new capability. And we’re finding a fantastic potential to merge a lot more and much more presents collectively.

If you’ll recall, I advised that Dell look at five companies — Appistry, Aster Information Programs, DynamicOps, Joyent and Univa — that can assist it add value to server sales either as standalone products or as deals focusing on particular industries or workloads.

Of individuals, Aster Info seems to be like an even much more-promising alternative, not only simply because Dell currently resells it as portion of a prepackaged analytics offering, but also due to the fact HP just obtained its data warehouse act with each other by acquiring Vertica. Addressing huge info requires, at the minimum, acquiring a merchandise that can supply sophisticated analytics versus massive amounts of info. IBM, HP and Dell-associate-turned-enemy EMC all obtained massively parallel data warehouse vendors not too long ago, and Oracle is offering its personal mega analytics appliances built atop its new Sun-acquired hardware. If Dell wants to play with the enterprise large boys heading forward, it just has to stake a presence in the big data marketplace, and Aster Data is each small enough and capable enough to do the task. For much more on Aster and Dell, arrive see Mayank Bawa, CCO and Co-founder of Aster Data, chat at our Big Info function in New York on March 23.

As I explained last month, although, the other 4 organizations also would meet Dell’s stated ambitions of buying small firms to expand its burgeoning programs administration company into new locations. All of them, as well as Aster Information, should seem even more appealing soon after taking into consideration how servers and enterprise sales have been Dell’s quickest-expanding revenue bases 12 months more than 12 months. All my possible Dell buys exist for the sole goal of helping enterprises manage big numbers of both bodily and virtual servers, and have proven themselves in use circumstances ranging from significant-scale net applications to high-efficiency computing, in industries ranging from on the web gaming to intelligence to financial companies.

Dell is in an intriguing place correct now, as it attempts to strengthen its cloud computing story, although at the same time getting to be a more popular alternative amongst large enterprises. It can’t manage as well many more large acquisitions like Compellent, nor ought to it attempt to completely realign its business model by getting AMD, but there are some excellent targets out there that can aid Dell kill two birds with one particular stone. Dell seems to get this by expressly mentioning application targets that add value to its core server enterprise, so we’ll see how it improvements on this mission.

Picture courtesy of Flickr user DRB62.

Connected material from GigaOM Professional (sub req’d):

  • The Info Center Is the New Box. Are You Ready?
  • Why Dell’s Cloud Computing Prospects Are Powerful
  • Think Converged Infrastructure Means Lock-in? Feel Once again.

Cloudera is happy to announce the fourth beta release of Cloudera’s Distribution of Apache Hadoop version three — CDH3b4. As normal, we’d like to share a handful of highlights from this release.

Since this will be the final beta before we designate CDH3 stable, our focuses for this release have been on balance, security, and scalability.

read more

Some about :

Read more:

  • cloudera theme for wordpress
  • how to balance cloudera
  • jambuddy (beta)

It looks like Eucalyptus’s enterprise proofs of concept are starting to pay off with customer wins; Tuesday morning, it’s athletic apparel brand Puma coming out as a customer, running Puma.com and related sites on a Eucalyptus-powered infrastructure. I noted yesterday that actual adoption of private cloud computing for production applications is still relatively low outside the service provider community, so Puma’s decision to come out as a cloud user is kind of refreshing. That it’s running a web site on Eucalyptus infrastructure might also help suggest what the first round of private-cloud-hosted applications will be.

Eucalyptus, has been the subject of some skeptical press since its semi-public split with former flagship customer NASA and the advent of OpenStack, but perhaps reports of its demise were greatly exaggerated. CEO Marten Mickos has exuded nothing but confidence in the company’s business model, pointing to large numbers of free downloads that could translate into paying customers, and the past few months validate his stance. While most private-cloud vendors make news with funding and new features, Eucalyptus has done so with two well-known customers: first, hotelier InterContinental Hotels Group and, now, Puma. Other private-cloud startups might have Amazon Web Services in their DNA, but Eucalyptus’s head start in installations and customer engagement might now be starting to pay off.

That Puma is using its cloud infrastructure for web sites also is telling. Using its existing infrastructure, Puma says it can handle traffic spikes for its primary website, as well easily provision new resources for “mini-sites.” With a few notable exceptions, public-cloud adoption appears to be largely centered on web hosting, so why should private cloud adoption be any different? If the underlying value propositions of flexibility and on-demand provisioning for users remain the same, it seems only natural that private cloud adoptees — whose primary contention with public clouds appears to be data security and not the delivery model — will follow the same path. Even with the added security of having cloud infrastructure in-house, moving legacy mission-critical applications to a new platform is no easy, nor necessarily desirable, task at this point.  Another publicly-named Eucalyptus customer, the U.S. government, runs USAspending.gov on a Eucalyptus-based cloud.

Perhaps I’m reading too much into the Puma news, but there do appear to be trends shaping up around Eucalyptus traction and web hosting as an emerging private-cloud use case. With true private clouds arguably a later creation than true public clouds, it only makes sense they might follow similar adoption patterns, with production applications coming on board only after a couple of years to experiment, and with web sites leading the way.

Related content from GigaOM Pro (sub req’d):

  • Defining Internal Cloud Options: From Appistry to VMware
  • Why OpenStack Has Its Work Cut Out
  • For Open Cloud Computing, Look Inside Your Data Center


Read more:

  • intercontinental hotels cloud computing
  • why intercontinental use cluod computing

3ones is a products advancement company for hire. They acquire merchandise for communities, buyers, and corporations. Their items seem on Internet websites, social networks, desktops, and cellular gadgets. They designed CompleteGuides.net – a destination for technical books.

Building on a MindTouch-driven publishing platform has enabled us to not only minimize production time by 350%, but to produce a social expertise. Volunteers have aided us make greater publications by proofing, recommending content, and sharing the site with their social networks.

Patricia Forest, Co-Founder, 3ones

Goal

3ones endeavored to open a new chapter in guide publishing for the technical market place. They set out to challenge the classic publishing model for technical writing. In speedily developing industries, publications for dummies and how-to guides are as several as 6 months out of date prior to they even hit the shelves.

Solution

3ones3ones chose MindTouch for its price and ease-of-use. Genuine-time collaboration involving authors, editors and graphic designers grew to become feasible instantly, with no formal consumer instruction. Single-sourcing the subject material and publishing across many channels concurrently adds worth with out adding energy.

Outcome

Employing MindTouch Technical Communications Suite, authors, editors and graphic designers have been capable to collaborate in actual-time, pushing their first guide to print in a matter of weeks rather of months. In the 1st 6 months considering that it was published, they’ve averaged over a thousand orders a month. Capitalizing on the social, viral facet of the platform, their order volume on a single marketing day was so wonderful that it brought on Google Checkout to crash.

Genuine Success

Prior to utilizing MindTouch, 3ones invested 200 hours of guide manufacturing time (excluding writing time) using a mix of MediaWiki, Google Docs and MS Word. With MindTouch TCS, this energy was reduced to 60 hours (thirty% of the time). These reduced production fees permit 3ones to market their publications (the two eBooks and standard print) at a third price of traditional publishers. Long run editions of their publications will get only ten-twenty hrs to create, thus adding incremental value about the lifetime of each and every edition.

Read more:

  • mindtouch commercial license hack

Web development nowadays plays a significant role in mature business processes. A major focus of especially work is creation of WEB sites or as it is called - the Web design. Every year more and more companies and individuals decide that it's time to get his own website. From the one point, website development by you, not a web development company at first glance may seem cheaper. But only at first glance. As in any business, development of WEB sites should be carried out by experts with a certain experience. In other words, everyone should do their job and often development your own web site only leads to a loss of time, not bringing the desired results.

Visionary leaders are well aware that it is necessary to trust the professionals in the sphere – the real web development company. In the following paragraph we are going to analyze more the tasks of both parts: the client and the web development company and to suggest what should be done when starting to create your own website. First of all it is necessary to set up the definition of web-development to make our analysis more clear. Web-development all in all is a very broad concept that includes the implementation of any network technology. Programming of the applications to work with clients through a web site – is a work for web development company. Creating a Flash movie for web site - it is also web-development for real. Not all the web-development involve working directly at a web site, there is a huge market of online advertising, software, etc. Let us consider the basic steps of developing the site as one of the most common tasks. It is assumed that we are qualified only in the area of the future web site - its subject matter. So the first thing we need - to find a studio or agency for which the web-development - the main type of activity. If a developer is defined, then plug the price determined - how much money we spend, depending on the services need that can be detected while working on the Terms of Reference (TOR) with the web development company.

The task of the developer is to offer us a set of services and structure of the web site. Our task - to analyze information and make a decision. Following the next step, when working on the Web site content the task of the developer - to make a specification for text and graphical content Web site with the objectives of the web-project and its further advancement. And our task in this case is to make available all the necessary information to the second side. It is important to know, that Web-development will not go on TK, if we do not approve it. If we approve it, without going, you run the risk of pointless money spending. TOR for development of the site should include a description of the design, structure, services, software modules, as well as the overall concept of the Web site logical to both you and web development company. This is the main document, in terms of further promotion of the website. Moreover, once the web site will be laid out, make changes in its appearance will be much harder.

Throughout a mobile phone phone this early morning, Rackspace Cloud President Lew Moorman flatly dismissed allegations that his firm bought Anso Labs to gain further energy in the OpenStack local community, and acknowledged that Rackspace is in talks with Microsoft about supplying a managed model of Windows Azure. Rackspace has made the headlines two times this week — first for getting Anso Labs, a San Francisco-based mostly consulting agency that assisted NASA create its Nova cloud-computing software program and launch its Nebula cloud, then for topping $ a hundred million in cloud income in 2010 — and Moorman addressed each in buy.

On Anso Labs and OpenStack

The Anso Labs deal is quite notable, in fact, due to the fact Anso was the brains behind Nova, the computing application that runs NASA’s Nebula cloud and that comprises the compute component of OpenStack. Bringing it in-home probably will support enhance the method of building that engineering, as well as the process of integrating it with the OpenStack Storage part originally created by Rackspace. Issues arose, nevertheless, when somebody started counting seats and realized that by buying Anso Labs, Rackspace now has three of 4 seats on the OpenStack governance board and 9 of ten seats on the Project Oversight Committee. Since OpenStack is an open source neighborhood, this kind of a sudden shift in energy stoked fears that Rackspace purchased Anso Labs so it could push its own agenda inside OpenStack.

Moorman says there is “no fact at all” to these allegations. In fact, he extra, the worries are acceptable, and Rackspace will hear to them before, possibly, suggesting changes to the governing board. “If we conclude up starting to be a dictator in this point,” he defined, “it won’t go everywhere.” That’s why Rackspace has but to deny membership to any contributor or to reject anybody’s code. Further, he noted, that with significant vendors such as Citrix, Dell and Cisco getting much more concerned with OpenStack, Rackspace wouldn’t most likely be ready to rule the venture with an iron fist even if it desired to. The “community is heading to keep on to hold us to account,” he stated.

That getting stated, Rackspace does intend to make money off of OpenStack. It won’t do so by utilizing Anso Labs’ experience to develop a Rackspace-only edition of OpenStack or a premium version that it could offer to third parties, but Moorman explained Rackspace is “working via” how it might give support providers to organizations deploying OpenStack. Anso Labs presents some interesting capabilities in phrases of real deployment help, Moorman famous, and there’s often Rackspace’s trademarked “Fanatical Support.” Yet another selection he suggested could be offering Cloudkick’s monitoring and administration computer software to OpenStack end users. Rackspace will be “much far more explicit” about its OpenStack assistance options in the up coming month or so, but, Moorman clarified, adoption of the software program is still its No. 1 target, and Rackspace won’t compete too aggressively towards its group partners.

On Driving Far more Cloud Revenues

Whichever income Rackspace is in a position to recognize from OpenStack support will just be more padding on the “cloud revenue” line of its quarterly earnings statements. Cloud revenue is made up of Cloud Servers, Cloud Files, Cloud Internet sites and Rackspace’s managed e-mail offering (although Moorman stated the former two drive it) and it topped $ one hundred million in 2010. Moorman said Rackspace isn’t seeing any decline in curiosity for cloud companies, and it only has plans to add more possibilities to its cloud portfolio. He hinted at forthcoming PaaS and/or SaaS offerings, stating that Rackspace certainly will move up the stack, as properly as carry standard hosting providers these as security and load balancing to the cloud. Astonishingly, Moorman also stated Rackspace is “actively engaged” in conversing with Microsoft about providing a managed model of Windows Azure, though there are no strategies in place nevertheless.

In the finish, though, Moorman does think OpenStack will be integral to Rackspace’s cloud accomplishment, even although the organization has done just fine with out it. The cause is that he views cloud computing as an ecosystem-driven company, a theory bolstered by the rising footprints of cloud providers like Amazon Net Solutions and VMware. He thinks OpenStack can serve the position of the cloud’s open ecosystem, and that Rackspace will reap a goodly quantity of organization from getting aspect of it. And there’s a great deal of enterprise to be had referencing Man Rosen’s month to month rely of which cloud providers are internet hosting the biggest amount of significant world wide web web sites, Moorman famous that the top 6 cloud companies put together to host less than two % of the prime 500,000 web sites.

If Rackspace emerges from what’s now a mini telco acquiring spree impartial — a likelihood that seems progressively possible provided its quickly-growing organization and penchant for making its personal acquisition lately — it seems poised to lastly make a run at Amazon Web Solutions. If not in customers, in income, but not with no a thriving OpenStack ecosystem.

Relevant subject material from GigaOM Professional (sub req’d):

  • Need to Cloud Companies Adhere to Amazon Down the PaaS Route?
  • Why New IaaS Companies Enter at Their Individual Threat
  • Is AWS On Track for $ 500 Million?

Read more:

  • does rackspace make money from openstack
  • openstack revenue

I really like to meet with commence-ups. There's usually energy, enthusiasm, creativeness and a desire to alter the entire world in some fascinating way. About the many years I've also examine and reviewed a lot of of the apps for the AWS Start off-Up Challenge.

read much more

The growth of unstructured data has outstripped structured data (in data bases) by almost 100 Xs, and this exponential growth rate shows no sign of slowing.

All of those files, photos, videos, documents, music and emails that we put up into the cloud on a daily basis - contribute to this deluge. The question is where to they go, and how are the cloud services companies dealing with and preparing for this mountain of data?

To a large extent the answer to this question lies in something called object storage. Object storage is a method for storing arbitrary collections of data. Rather than doing so in a traditionally structured form, object storage is non-hierarchical, using a flat address space.

Unlike traditional storage which uses physical location to find and retrieve data, object storage uses a unique identifier which allows users to access data with no knowledge of its physical location. In this way multiple copies of the data can be created, and moved within the system, even between data centers, seamlessly. This allows object store systems to grow and to change their internal mapping and configuration without impacting applications and users.

An object has both data (an uninterrupted sequence of bytes) and metadata (a somewhat extended set of attributes describing the object, as well as other system or business specific attributes such as replication rules or protocols).

An object is referenced with a unique identification key, typically alphanumeric characters or a very large integer.

A software layer commonly sits in front of the storage back end, and behaves as a broker, connecting the application(s) and storage infrastructure. The languages that are generally used as the interface between these layers vary. They include a variety of proprietary APIs, industry specific syntax like DICOM (for the medical records industry) and web protocols such as a RESTFUL / SOAP interface. HTTP/1.1 being the most used protocol.

Object storage systems provide a simple resource oriented interface to client applications which enables a new class of storage virtualization. Object storage systems could be considered to be the only natively cloud based storage systems available. Designed specifically as distributed systems over many servers and potentially multiple data centers. Data is replicated several times so that the loss of any given server does not lead to a loss of data.

A new generation of object storage, dubbed ‘organic storage’ (a reference to organic computing) further emphasizes the autonomic characteristics of object storage:

Organic storage includes dynamic traits such as self-healing and learning protocols that facilitate automation by allowing the system to be adaptive to changes in its environment (e.g. auto migration and replication), adaptive to changes in human needs (e.g. load balancing and auto tiering based on actual data usage).

This radically simplifies operational and client – server interactions with and within the system, lowering cost, risk and management overhead.

Object storage was initially designed in the late nineties in the form of CAS (Content Addressable Storage) for backup. Towards 2005, Google, Facebook and Amazon had to solve the problem of storing billions of files for their consumer services, and through R&D each came up with a form of object storage.

Initially these early generations of the technology were unsuitable for high performance primary applications, due to low transactional performance; however with the advent of Organic Storage, performance has radically improved, to the point where it is now equivalent with or even superior to traditional NAS or even SAN systems.

Uniquely suited to very large scale (petabyte) storage of unstructured content, Object storage is widely recognized as the ideal solution for storage vitalization / archiving and web 2.0 user generated content applications. With the advent of high performance organic systems, though object storage has been reclassified and is increasingly finding proponents among high performance primary applications such as the email storage industry.

An example of Organic Storage is Scality’s RING Organic Storage. Unlike traditional object storage RING is flexible enough to accommodate both the low performance, data longevity requirements of backup and archival use cases, as well as the high performance needs and support for very large numbers of files, which are required by email services and Web 2.0. It even accommodates the large file sizes, very high IOPs needs and security requirements of surveillance (CCTV) storage.

RING Organic storage offers a robust, flexible way for large cloud storage providers to accommodate the many millions of objects currently stored, and also provides an easy and cost effective way to scale to many billions of objects as the need arises.

Monique Shefer is a technology analyst and strategic consultant to the software and software as service industries. Shefer is currently working for Scality – Storage System Pioneer and developer of RING Organic Storage.

« Previous PageNext Page »