Posts tagged ‘waiters’

With Blue Insight, IBM proposes the largest cloud priv? existing

Technology – the American giant has advertisement the startup of his infrastructure of cloud computing baptisee Blue Insight able to store a petabyte data. Will Blue Insight be D? does access use in-house by the 200.000 employees D? IBM.

If IBM N `is not gets into the first train of the cloud
computing, the American firm put the means to make up for its lost time. Big
Blue has just presented its cloud deprives which is quite simply largest
ever conceived.

to give an order D `idea, IBM indicates that Blue Insight, C `is
its name, can contain on its waiters the equivalent of 250 billion
pieces of music. For the moment the cloud will be grinds in-house by
200.000 pay group which will reach has data stockees on one
hundred waiters. The system is built on the IBM waiters of the range
System Z and uses the solution of rachetee for 5 billion dollars in 2007.

IBM counts
then to market a service baptizes Smart Analytics Cloud which will propose
in particular the estimated ones on the incomes, the sales quotas, the ruptures
stocks and the levels D `inventory. (Eureka Press)

January 14, 2010 at 11:43 am Leave a comment

C2S reconciles storage and virtual machines

tude of case – the subsidiary company of Bouygues has associ its projects of virutalisation of waiters and work stations, an architecture of storage which combines dduplication with the level of bays and tools of rplication and Thin Provisioning de VMware.

Filire data-processing of the Group Bouygues, C2S is a socit data processing departments which accompanies the DSI by the group within the framework of projects of recasting, of modification and D `volution of their infrastructures. C2S dveloppe and forms the others galement entits of Bouygues of the mtiers applications.

In 2007, the socit starts a plan of rationalization which results in an important recourse to. The waiters of the company are the first tre ports on virtual machines, that is to say about fifty waiters. C2S does not count thus dsormais more but five physical waiters and two ESX 3. 5 of VMWare (and two ESX supplmentaires for the redundancy).

L `storage space divided utilis by 7 grce the dduplication

Within the framework of its mission of formation, it undertaken has then engag a project of virtual work station (VDI) for its three classrooms. For, C2S S `supports on customers lgers Wyse (15 terminals by rooms) and the solution View 3 of VMware.

« Previously, before each training course, one needed remasteriser the work stations, which ncessitait a journe man. With the solution of VDI, for prparer the virtual machines and the mtiers applications to which the trainings relate, this tche is excute in one hour » explain, Pascal Bouthelot, chief of the service equipments production at C2S.

a bay of 4 To for the storage of the images

However, such an architecture (stations and waiters virtualiss) ncessite also a storage solution adapts. Give associate with the virtual machines are store on a bay of storage NetApp (in cluster) FAS 3020 C of 4 To, with the active one. This technology allowed C2S to divide by seven storage space utilis.

For rpondre its increasing requirements in terms for volumtrie, C 2 S.A. acquired one second NetApp bay, galement of 4 To. If the premire is out of Chanel Fibre (with discs of 146 Go 15.000 tours/s), the bay supplmentaire is as it in protocol iSCSI, but hberge of less critical volumes and less consumers.

« With the dbut of our project, there is a little more than two years, we have test. With the poque one, it conceals more constraining D `technically to associate this protocol of the virtual machines in VMware VI 3 – management of the Frame Jumbos and the almost impossible MPIO » , Pascal Bouthelot justifies.

a recourse to the native functions of VMware to optimize the coteuse licence

In S `pressing on Mrs. technology of storage, C 2 S.A. galement test, in environment of production, a recovery plan D `activit with a distant location. The objective: to quickly allow, in the event of disaster, the redmarrage of its infrastructure composes D `a hundred of virtual machines.

For that, another bay has T connects to the principal site by black fibre. On the level of the software tools, C2S S `is appuy on SnapManager for Virtual Infrastructure and SnapMirror for the rplication, remplac since by VMware SRM and the fonctionnalits of high disponibilit like Fault Tolerance.

« Aujourd `today empite more and more on fields before dvolus with the manufacturers. Thus, the ditor proposes dsormais function …

September 6, 2009 at 11:31 am Leave a comment

Olivier Parcollet With the consolidation, consumption dnergie has T divides by four

Opinion D `expert – Stages of the virtualisation of waiters and the work station, impact of lenvironnement virtual on storage, the band-width and ladministration with the daily newspaper: larchitecte systmes and rseaux division its exprience.

Quand do you have commenc you intresser the virtualisation?
We are penchs on the virtualisation there is bientt seven years because D `a problme of prennit of our technical platform. This one, whose cycle of renewal is long, of the 10 15 years order, hbergeait critical applications functioning under Windows NT4.
# Our infrastructure conceals obsolte and it conceals impossible D `to us to plan to carry NT4 on waiters rcents. Rcrire the applications, when C `TAIT allowed, would have taken several months, even annes. The only alternate one thus conceals to find a solution enabling us to free itself from this problmatic matrielle.
# We have virtualis a premire application with VMware Workstation, in dpit of the recommendations of the ditor. We are thus not D `a not very powerful cluster NT4 a stable systme and which bnficiait directly of the power of processors of the new matrielle infrastructure.
# A few months later, we have dcid D `to apply Mrs. receipt, under VMware Server this time, all us waiters within the framework of their renewal. is
# Comment grs the diffrents virtual waiters of Setao?
Aprs the first project of virtualisation, we laid out D `about thirty VMware waiters, but without comprehensive view from what occurred on these machines. An application of total supervision conceals ncessaire.
# We have tudi Virtual Center then. But the cot appeared prohibitory to us since for only a few thousands D `euros moreover we could accder a version suprieure of the application of supervision: VMware ESX. C `is thus this solution which has T retained. The machines virtualises, 63 on the whole, have T import in ESX. The opration S `is made distance in one weekend.
# Of how much physical and virtual waiters is composed dsormais your infrastructure?
70 virtual waiters are currently in production. This total will pass bientt a hundred. As for the physical waiters, they are passs of 30 3. The infrastructure has T double for questions of continuit D `activit. In all, we thus count six rpartis out of two farms of three waiters.
# a farm is active, and the other is place in stand by with rplication on a site of help connect out of fiberoptic, and situ 19 kilomtres of our computer room.
# is the virtualisation concrtement source D `conomies?
the advantage is not done on the level of the cot D `acquisition, almost quivalent that D `a made architecture of physical waiters. A company should not think qu `it will gain there on the price D `purchase. Virtualisation N `is not an investment of short term.
# the vritable asset is at the level of the administration, grce in particular a provisionning of the waiters the request. Provsionner a waiter supplmentaire rclamait before between one and two days. Rcemment, J `have dploy two Linux waiters in approximately 15 minutes.
# is the performance altre by the installation of the virtualisation?
the systme is completely stable and almost indestructible in our configuration. Lorsqu `a physical waiter falls, the virtual machines are automatically semi …

July 6, 2009 at 3:55 pm Leave a comment

adjusts its infrastructure of storage the virtualisation

tude of case – to ensure the safeguard and the rplication of give its hundred virtual waiters, consolids on 6 physical, bioMerieux waiters set up a robust infrastructure of storage structures in MetroCluster.

Spcialiste of in vitro diagnosis in the field of the infectious illness, cancer and the cardiovascular urgencies, bioMrieux east prsent in 150 countries and account more than 6.000 collaborators. The company franaise has ralis in 2008 plus D `a billion D `euros figure D `businesses. Its activit, subjected constraints rglementaires, S `supports on an important computer infrastructure.
# In France, bioMrieux laid out thus into 2007 of more than 250 waiters, in major hbergs part on its site central. Moreover, this park of waiters grew D `around fifty D `units each Anne. The data-processing Management of the group begins then S `intresser the virtualisation, for rpondre a need for consolidation, but not only.

100 virtual machines on a cluster of 6 waiters

« We have dbut the day before on this technology in order to find a solution being able to offer amliorations to us on several axes: the scurisation, the consolidation by the rduction of the number of physical platforms, the fall of the nergtic invoice and the space saver in our center of give. The flexibility in the placement of new waiters, the request of the mtiers, galement conceals a central axis of the &amp project; #187; rsume Andr Ferlay, responsible infrastructures systmes & rseaux for bioMrieux.

structures It of virtualisation and its infrastructure of storage associates owe rpondre several requirements, among which the maturit of the solution, the prsence D `options of scurisation like the recovery heat, the fonctionnalit of snapshot on the level of storage and the possibilit D `hberger a great number of virtual machines. bioMrieux a ratio of 1 per 20 aims indeed, C `is–to say a physical waiter for 20 virtual.
# At the end D `a premire tude of the commercial offer, VMware and NetApp are retained. Intgrator APX is slectionn to set up structures it being composed, for the virtualisation, of VMware ESX 3. 02, and of bay FAS 3040A of NetApp for the storage part.
# « A file server of NetApp implment had dj T in 2007 on one of our sites for rpondre the needs of his 800 users. We had been able this occasion to test of them the fonctionnalits rplication, of snaphot and the tools D `administration. The product of NetApp prsentait moreover the advantage of the adaptabilit VMware and offered several modes of connection: Chanel fibre and iSCSI » prcise Sbastien Regnier, architect systmes the bioMrieux one.
# This premire ralisation in matire of virtualisation thus results in the installation of six physical waiters (in cluster) hbergeant 100 virtual machines, and connects in iSCSI the bay of storage (two ttes) D `a volumtrie of 8 To of give. (continuation…)

Page 1 |


July 5, 2009 at 7:58 am Leave a comment

Atari simplifies its data processing and rduit its cots

Technology – Within the framework dun project of simplification of sound systme dinformation, the socit of play vido Atari has consolid its waiters with the virtualisation, aprs however a tude dligibilit. Profit: an operating burnup in fall of 50 70%.

Subsidiary of the group Infogrames, produced Atari, said and distributes plays in more than 60 countries. Jusqu `in 2007, this pioneer of the play vido S `supports on two autonomous data-processing directions having four datacenters: two in the United States and two in France.
# This mode of governorship S `relatively coteux avre and results in the volont D `to standardize the systmes D `information. Objective of the project: to simplify it data-processing of the company and thus rduire its dpenses of operation.
# « The key question of this project of simplification conceals: which solution for rduire cots, amliorer the robustness, in particular on the part recovery plan D `activit, and accrotre the flexibilit of the systme D `information? We had dj effectu regroupings of fonctionnalits on waiters » explain the Defect Prsident Management Information Systems D `Atari, Nicolaas Laarman.

an operating burnup rduit of 50 70%

the rponse then considers: the virtualisation. A premire tude in-house ralise allows the data-processing direction to determine the contributions of this technology. This slap pralable dbouche on a tude D `ligibilit intends to identify the physical waiters being able tre virtualiss, with like critre central the rate D `use.
# Atari chooses the platform of virtualisation of VMware and PlateSpin for the aspiration of the configurations of the waiters. « Through diffrents tests and return D `users of VMware, the solution appeared to us in June 2008 like more robust than the other solutions of the march. The simplicit D `administration, the compatibilit with diffrents OS, the management of the clusters and the price galement have motiv this &amp choice; #187; dtaille Nicolaas Laarman.
# the intgrator Dell must raise a dfi: to buckle the installation of the new architecture in three months. Indeed, about thirty waiters D `Atari arrives at the end of the lifetime or at the end of their leasing agreement. The company thus wishes to consolidate these physical machines and thus to post a return on investment quickly.
# Atari waits moreover of Dell qu `it ensures the transfer of comptences so that its quipes data-processing can manage they-Mrs. the virtual infrastructure at the end of the three months. Four top-of-the-range physical waiters in cluster thus will accomodate more D `about thirty virtual waiters with like bnfice of gnrer a rduction the operating burnup (cot total of possession) 50 70%.

Still few visibilit on the virtualisation certain applications

Atari can moreover consider dsormais rduire the number of its centers of give of four two. A consolidation which will result in a fall of its dpenses into hbergement and in lectric consumption.
# But the virtualisation ncessite however to take certain prcautions. « It is necessary to be shown to be attentive with the needs for its waiters in terms of performances, of disponibilit in CPU and D `enter-exits on the hard drives. A tool of business intelligence for example needs resources. The console D `administration of VMware allows you dpla …

July 5, 2009 at 6:47 am Leave a comment

Sun Microsystems dvoile 7 new waiters quips in Intel Xeon 5500

Technology – the manufacturer amricain comes his turn from dvoiler a range of low waiters on the new processors Nehalem dIntel. The seven modles are rpartis on three families: Virtualisation, Company and Internet, and High efficiency.

Aprs HP, IBM, Dell, and
rcemment, Sun Microsystems has dvoil its range of waiters dowries of
processors D `Intel.

a rpartie offer on three families of products:
Virtualisation with Sun X4270 Fire and Blade X6270, the Company and
L `Internet with the Fire X4170 X4275, X2270 and the work station Ultra 27 and
finally High efficiency with Blade X6275.

These seven waiters have in common
to propose a multi-layer storage SSD/flash/SAS/SATA. The use D `one
moired flash ddie nonremovable east intends liminer the narrow parts
D `tranglement for the accs with give.

Sun announces the disponibilit immdiate
of these products. (Eureka Press)

April 29, 2009 at 7:40 pm Leave a comment

Google lve the secrecy on its centers of give

Google always has T miserly D `information concerning its centers of give. The glove of the Web holds thus secret the exact position and the number of the latter. With the occasion D `a confrence in the United States, Google has nanmoins rvl some information about these famous centers of give.
# the group thus admitted conceiving those by connecting diffrents containers dj quips their waiters and of their quipement of cooling. This architecture east exploits by Google since now 2005.
# Each container can accomodate a little more than 1000 waiters. Cooling is assur grce a systme D `water rfrigr. The waiters are thus fixs with the ceiling of the container, allowing water to circulate below, then galement through the diffrents racks.
# In its new center of give 7000m2 baptis Data Center has, Google has install in all 45 of these containers. The rvlations of Google are not without pointing out the project Blackbox de Sun which also proposes him to provide containers with their computer infrastructure intgre.

April 14, 2009 at 11:31 am Leave a comment

Older Posts

December 2019
« Jun