[data fragments]

EMC and SAP plan a romp in the clouds

The Register writes Today:

"EMC may partner with German software giant SAP to develop a suite of web hosted storage and application services.

In an interview with Reuters, SAP executive Doug Merritt said the two companies are in early-stage talks to add SAP's business applications to EMC's burgeoning software-as-a-service (SaaS) storage platform.
Both companies are mum on any technical or financial details of the proposed collaboration. But let's look at how things lay.

In January, EMC unveiled MozyEnterprise, the first offering in its new hosted software line dubbed "Fortress." The online backup and recovery suite is based on technology acquired from Berkeley Data Systems last October. EMC made sure to point out Mozy is only the first step in expanding the SaaS operation.
SAP has also been noisy about launching its own flavor of on-demand service called BusinessByDesign to combat Salesforce.com. At an announcement in September, SAP said it will look to partner with companies to provide applications and hosting.

The two puzzle pieces could logically fit together. As could be expected for an "early-stage" corporate affair, both companies declined to comment. But EMC veep Chuck Hollis helps push speculation. In his company blog, we see Hollis resist offering details, but perfectly happy to paint a picture of how such a collaboration would be, oh, such a grand idea.

Sounds a little like suit-speak for 'announcements are upcoming.' "

Better Search for E-mails?

"IBM on Thursday introduced a test version of a semantic search engine for companies that want to give employees more advanced tools for searching their e-mail stored inMicrosoft Outlook or IBM's Lotus Notes.

The OmniFind Personal Email Search, developed by researchers at IBM facilities in California, Israel, and India, takes search beyond keywords by being able to make associations between the underlying concepts of words often used in corporate e-mail. In doing that, IBM is looking to offer customers technology that can help them retrieve useful information hidden in e-mail databases.

For example, if a person is looking for a colleague's phone number, then typing "John phone" in the query box would return John's phone number. That's because the system is smart enough to make the association that the person is looking for John's number, and not just any phone number in an e-mail with the word John in it.

To do this, researchers built an index of keywords found in corporate e-mail, and then built another index of associated concepts and relationships, Shiv Vaithyanathan, manager of unstructured information mining at IBM's Almaden Research Center in San Jose, Calif., told InformationWeek. When a query is submitted, the system first matches the words with those in the keyword index, and delivers results based on the associations. Researchers have added rules to help the system determine what information is most likely being sought.

OmniFind Personal Email Search uses the Unstructured Information Management Architecture, an open source software framework for semantic search. IBM was the developer of the technology, which is now under the Apache Software Foundation."

[source: here]

Ist Dell wirklich bereit für Green IT?

Computerzeitung online hat am 27. 11. mein Interview mit Dorothee Stolzenberg, Geschäftsführerin von Dell in Frankfurt, veröffentlicht. Dell reklamiert einen "ganzheitlichen Anspruch" für sich und möchte Green IT nicht nur bei sich selbst verwirklichen, sondern auch bei Lieferanten und Kunden auf diesen Pfad bringen. Doch was ist "ganzheitlich" und wie passt "green" zu einem amerikanischen Unternehmen?

More: here.

The Data Center Race

Nicholas Carr, always on the run to find new examples for his theories about the big change of IT, also points on this:

"In a further sign of the vast scope of the battle to dominate the emerging utility computing grid, Microsoft today announced that it plans to build a large data center in Siberia, notes Data Center Knowledge. According to a local press report, the center will be located in Irkutsk (at a site to be determined), will hold "tens of thousands of servers," and will cost approximately $500 million to build and outfit.

The Siberia announcement is another sign that Microsoft is picking up the pace of its buildout. Earlier this month, the company confirmed that it will construct a "mega data center" near Dublin, Ireland, and also announced that it will build its largest data center yet - the size of eight football fields- in the Chicago suburb of Northlake. These centers, all dedicated to supplying the company's Live-branded software applications over the internet, supplement server farms already under construction in Quincy, Washington, and San Antonio, Texas.

Microsoft is playing catchup with Google, which is already reported to have dozens of data centers throughout the world and has this year announced new centers in North Carolina, South Carolina, Oklahoma, Iowa, and the Netherlands. (Today's Des Moines Register has a long article with some new details on the Iowa plant, which Google appears to view as an important "hub" for its North American network.) Altogether, the new data centers announced by Google and Microsoft this year represent capital investments of more than $4 billion.

Both Microsoft and Google are looking for creative ways to reduce the electricity consumed by their big computing plants. Google located its center in The Dalles, Oregon, beside the Columbia River in order to use the river's water to cool its computers, reducing the need for air conditioning. One of the reasons Microsoft chose the Chicago area for its largest center is the cold climate. A Microsoft spokesman said "the cool weather can be tapped in innovative ways to cheaply and efficiently cool down racks and racks of tens of thousands of computers that generate loads of heat." One assumes that Siberia holds a similar attraction.

If local communities are hoping these giant centers will create a lot of new jobs, they'll probably be disappointed. Microsoft has revealed that the Chicago facility, which may be the biggest data center on earth when it's completed next year, will employ a grand total of between 35 and 50 people."

[source: here]

Rumors: Is Google rolling its own 10 Gigabit Ethernet switches?

Network World writes:

"Google may be building its own 10 Gigabit Ethernet switches to support its massive data-center needs, according to a recent blog post by Nyquist Capital.
Andrew Schmitt, general partner of the investment firm, caused a stir when he suggested Friday that Google may be responsible for a shift in the market for optical components that could not be accounted for otherwise.
Schmitt wrote:
"We were watching shipments of SFP+ components for 10GbE in the market but simply couldn’t account for their end destination - sort of an optical component dark matter problem… Through conversations with multiple carrier, equipment, and component industry sources we have confirmed that Google has designed, built, and deployed homebrewed 10GbE switches for providing server interconnect within their data centers." "

[source: here]

A Perfect World

Noch ein Standard: Fiber Channel over Ethernet (FCoE)

Aus einer Presse-Miiteilung von IBM:

"San Diego, Kalifornien / 5. April 2007 - eine Gemeinschaft von IT-Anbietern hat die Schaffung einer neuen Technologiespezifikation mit dem Namen "Fiber Channel over Ethernet" (FCoE) dem T11-Komitee des US-amerikanischen National Standards Institute (ANSI) vorgelegt. FCoE würde es ermöglichen, SAN-Datenverkehr nativ über Ethernet-Netzwerke zu transportieren und gleichzeitig die Investitionen schützen, die Kunden bisher in ihre Speichernetzwerke getätigt haben. Der Vorschlag einer direkten Umsetzung von Fiber Channel über Ethernet hat die Unterstützung von wichtigen Industrieanbietern wie Brocade, Cisco, EMC, Emulex Corp., IBM, Intel, Nuova, QLogic, und Sun Microsystems. (...)

FCoE könnte Anwendern neue Wahlmöglichkeiten für weiter übergreifende Server-Konnektivität im Rechenzentrum bieten und würde heute existierende Konnektivität und Protokolle, die bei Speicher- und Rechenzentrumsvernetzung genützt werden, ergänzen."

"Ergänzen" ist vermutlich nicht ganz richtig. "Ersetzen" wäre zutreffender. iSCSI-SANs, die gerade im Kommen sind und in den USA schon einige Marktanteile errungen haben, könnten mit FCoE etwas Gegenwind verspüren.

Fehlt nur noch eine SME-Variante von Fibre Channel. Billig, einfach zu verwalten etc. ...

UPDATE: Was soll das Ganze? Hierzu zwei Antworten:

a) "FCoE: Run Away, Its the Monster

Today's ANSI news: .......... FC vendors propose FCoE. Fibre Channel Protocol over Ethernet or something like that. Serial SCSI running over Ethernet without TCP or IP. Let's see, that has been tried before with Nishan's SOIP, Adaptec's EtherSCSI and other less well known technologies - and killed off by whom? Most of the companies proposing FCoE. Ironic.

So, if you are a FC vendor and it looks like 10Gb Ethernet is going to be part of the tiger (along with iSCSI) that eats your lunch, what do you do? You resurrect old ideas that you hung five years ago, dust off the corpses, insert zombie brains and call it the new shiny FCoE (Frankenstein Convergence over Ethernet?). The only difference between what you killed yesterday and your unholy experiment today is that this bad boy saves your business. It doesn't really have that much to do with customers, except that you get to squeeze them a little longer."

[source: Marc Farley, EqualLogic]

b) "FCoE - Quick Take

I just read Marc Farley's blog on FCoE (FCoE: Run Away, Its the Monster) and felt compelled to write about this subject as a result. I really like Marc's passion but I can't say that I agree that FCoE is a monster. My initial reaction is that FCoE is a good thing.

FCoE - or FC over Ethernet allows you to use Ethernet to build FC SANs instead of - or in conjunction with FC infrastructure. The value of FCoE is that it perserves WWN addressing, zoning parameters, multipathing, SMI-S, etc already established with existing FC SANs. It actually makes it easier for FC-centric end user shops to leverage Ethernet. You will be able to use FC switches by adding FCoE ports. It will be Ethernet "in" from the host systems and then FC "out" to the existing storage systems that support FC ports. It allows the FC storage adminstrators to still be in control of their own fate versus having to go to the networking guys for infrastructure.

Additionally you be able to have FC and iSCSI traffic (as well as all of your other network traffic) sharing the same Ethernet infrastructure. That is a good thing. FC and iSCSI will be able to share the same front end of the network happily moving across common Ethernet networks. In fact, FCoE may actually help iSCSI gain a greater integration over time. FCoE can be a transition point from FC to iSCSI.

One of the big decisions that storage system vendors will have to make is whether they will support native FCoE ports in the future. If yes, then FCoE has a chance of being more viable over the long term. However, if the storage system vendors feel that the extra development and Herculean interoperability testing effort isn't worth it, then iSCSI will ultimately replace FC at the high end - just as it is doing in the mid-tier of the market.

It is important to remember two things. First, any new standard and technology takes much longer than anyone supposes to have a real market impact. That will be true of FCoE. Second, iSCSI provides greater benefits than just running over Ethernet. iSCSI and the IP protocol are much more virtualized than FC - making it much easier to manage and scale. FCoE does provide value by leveraging the existing FC ecosystem but it has nearly all of the limitations of FC: You still need specialized equipment; FC expertise; it is more distance limited than iSCSI, and it requires more manual management.

Like so many things FCoE is not a panacea but it could be a good transitional technology that helps FC-centric data centers save on some costs by sharing Ethernet infrastructure while perserving their FC investments. There are four possible outcomes for FCoE (that at least I can think of right now):

1. FCoE could be a non-event and gain no traction.

2. It could help pave the way for iSCSI to be dominate at the high end - FCoE would be transitional as a step towards iSCSI. It perserves the FC ecosystem but gets the data center sharing much of the same Ethernet infrastructure.

3. FCoE could replace FC at the high end if storage vendors embrace it and if there is still a bias against iSCSI at the high-end. This will be timing more than anything. As iSCSI becomes more mature, there will be less of an issue with adopting it at all levels. For FCoE to gain any real traction, the storage system vendors will need to support it quickly.

4. It will be the dominant SAN interconnect of the future. I wouldn't bet one dollar on this one coming true.

FCoE does support something that Ethernet inventor Bob Metcalfe said to another ESG analyst, Brian Garrett - "I don't know what comes after Ethernet, but it will be called Ethernet"."

[source: Tony Asaro, Strategy Enterprise Group]

Jobs and the 13.000-Dollar-PC

Apple just announced a four-core PC for $13.000. Steve Jobs respond to some criticism:

" How dare you rip us on our pricing?

Industry pundit Nick Carr is taken aback by the $13,000 price tag on our new top-end Mac Pro with an eight-core processor, 16 gigs of RAM and 4x750GB hard drives. "Didn't anyone tell Jobs that these things are supposed to be commodities?" he writes. (I can hear his hands wringing from all the way across the country.)

Not sure I get it. Oh wait. I guess $13,000 is a lot of money for some people? I don't know. It's been so long since I carried cash or actually paid for anything that I don't really have much sense of prices anymore. And living out here in the Valley where everyone basically has more money than they can ever spend, well it kind of skews your perspective.

Thing is, Nick Carr, here at Apple we don't build to price. And unlike a lot of other companies, we don't sit around trying to guess what customers want or -- worse yet -- asking customers what they want. God only knows what they'd come up with. Probably they'd just sit there shrugging and eating donuts in the focus group room. Fact is customers never know what they want. If they did they'd all be running tech companies right? They're looking to us to tell them what they want. That's what we're here for. That's our job. So you know what we do? We build what we think is cool. We build the products we'd like to own. And yeah, a $13,000 souped-up Macintosh is gonna be pretty popular in the centimillionaire-Silicon-Valley-engineer crowd. Trust me. We know what we're doing."

[source: here, 5-4-07]

Microsoft's Ozzie about Software Future

In an interview former Lotus manager Ray Ozzie, now with Microsoft, said:

"When we, as an industry, communicate the meaning of an architectural shift to customers, sometimes it's great to take an extreme position because it helps people to understand the benefit of this new era. In first generation "software as a service," people tried to push the browser as far as it could go. But the most important mission of vendors is to figure out what value they are delivering, not how they are delivering it. If you look at people like Salesforce.com, they may talk "software as a service" being [exclusively] through a browser, but they have an offline edition ... What we as an industry need to deliver are seamless experiences - however those things are accomplished - to do the appropriate thing in the browser and the appropriate thing on a laptop or on a device to solve that problem. So the way I view it is, first generation "software as a service" really just meant browser. Second generation means weave together hardware, software and services to accomplish a specific solution."

More here.

The Sermon of the Tapes

"State of the Art - The Sermon on the Tapes
By DAVID POGUE

Dearly beloved, let us bow our heads and mourn the passing of so many beloved media storage formats: Home-movie film reels (1932-1975). Floppy disks (1969-2002). Eight-track tapes (3:15 p.m. on Aug. 3, 1965, to around dinnertime on Nov. 5, 1988).

Let us also say a prayer for vinyl records, audiocassettes and VHS tapes. They may not be dead yet, but they do wheeze a bit going up stairs.

When their time comes, they, too, shall ascend to Format Heaven, where they shall be joyously reunited for all eternity with stone tablets, illuminated manuscripts and cave drawings.

But yea, though these recent formats have the approximate shelf life of milk, there is still time to resurrect them. Unto ye I say, copy these crumbling films and disks onto shiny new recordable CDs and DVDs while they still walk among men.

Here, then, for your clipping-to-the-fridge pleasure, is the latest wisdom on helping old recordings pass into their next life."

Here then.