Monthly Archives: mei 2013

De echte cloudexpert bestaat niet

31 mei 2013
By

Cloud Computing Consultant

Expert van Computable voor de topics: Cloud Computing en BPM

Meer

Nu er steeds meer voorbeelden komen van bedrijven die naar de cloud gaan heeft dit zijn uitwerking op de overige bedrijven. Bijna ieder bedrijf begint zich toch wel af te vragen of ze iets missen als ze niet de cloud op een of andere manier gaan gebruiken.


Als je er een Google sleutelwoord analyse op los laat zie je dat er veel gezocht wordt op ‘cloudexpert’. Laat ik duidelijk zijn. Er zijn geen cloudexperts, en degene die er dichtbij komen kun je niet betalen. Er zijn wel cloudgeneralisten, heel veel cloudgeneralisten. Van marketingmanagers tot adviseurs, iedereen kan wel iets over de cloud roepen. Maar laat ik eerst onderbouwen waarom er geen cloudexperts zijn.

Geen expert

De voornaamste reden is dat cloud het gehele it-spectrum raakt. Van netwerken, virtualisatie, infrastructuur tot het maken van databases, interfaces en de verbindingen er tussen. Dan is er nog security en beleid. Oh, en elke adoptie doe je graag onder architectuur of project management. Er is denk ik niemand op deze planeet die zich over alle aspecten expert kan noemen. Software is zo anders dan infrastructuur en netwerken. Als je kijkt naar het veel genoemde, IaaS, PaaS, SaaS, dan is de doelgroep ook anders. Infrastructuur heeft de grootste raakvlakken met systeem/netwerk beheer, PaaS bij ontwikkelaars, SaaS bij ontwikkelaars, gebruikers en business in het algemeen.

Dan is er nog de dimensie consument, mkb, enterprise. Consument zijn we allemaal, maar waar je vlieguren mee maakt is waar je referentie kader ligt. Zonder referentie kader geen expert.

Je bent dus niet op zoek naar een cloudexpert, maar naar een cloudgeneralist met wellicht expertise in de branche

Bron: Computable Lees het complete artikel hier: http://www.computable.nl/artikel/opinie/cloud_computing/4734181/2333364/de-echte-cloudexpert-bestaat-niet.html

Server racket ‘challenging’ in Q1, says IDC

30 mei 2013
By

It is not a fun time to be peddling servers, but it is probably a great time to be buying them, according to the box-counting wizards at IDC.

Both server revenues and shipments were down in the first quarter, marking the fifth of the past six quarters where revenues took a dip. There are so many different kinds of transitions going on in Server Land it is a wonder it isn’t worse, frankly.

In the quarter ended in March, IDC reckons that server revenues worldwide (at the vendor factory level, not at the retail level) came to $10.94bn, a drop of 7.7 per cent from the year-ago period. Shipments dropped by 3.9 per cent to 1.89 million boxes across all form factors, which include tower, rack, blade, and density optimized machines.

“Customer demand for new servers is being impacted by ongoing server consolidation, technology transitions, and challenging macroeconomic conditions across the globe,” explained Matt Eastwood, general manager of enterprise platforms at IDC, in a statement accompanying the ships and bucks of the server biz.

“In fact, every geographic region except Asia/Pacific experienced revenue contraction in the quarter. It is clear that challenging market conditions are increasing the competitive dynamics for server market share globally, particularly since compute represents a critical element of larger IT transformations that continue to reshape broader enterprise IT market opportunities.”

Gartner pretty much said the same thing in its Q1 server report on Tuesday, although it dices and slices the market slightly differently so you need to have a look-see at both sets of numbers to get a three dimensional view of the server terrain.

IDC concurs with Gartner: The server biz was challenged in the first quarter

By IDC’s numbers, as HP’s server business contracted by 14.8 per cent, nearly twice as fast as the market at large, but it was

Bron: The Register Lees het complete artikel hier: http://go.theregister.com/i/cfa/http://www.theregister.co.uk/2013/05/30/idc_q1_2013_server_numbers/

Server racket ‘challenging’ in Q1, says IDC

30 mei 2013
By

It is not a fun time to be peddling servers, but it is probably a great time to be buying them, according to the box-counting wizards at IDC.

Both server revenues and shipments were down in the first quarter, marking the fifth of the past six quarters where revenues took a dip. There are so many different kinds of transitions going on in Server Land it is a wonder it isn’t worse, frankly.

In the quarter ended in March, IDC reckons that server revenues worldwide (at the vendor factory level, not at the retail level) came to $10.94bn, a drop of 7.7 per cent from the year-ago period. Shipments dropped by 3.9 per cent to 1.89 million boxes across all form factors, which include tower, rack, blade, and density optimized machines.

“Customer demand for new servers is being impacted by ongoing server consolidation, technology transitions, and challenging macroeconomic conditions across the globe,” explained Matt Eastwood, general manager of enterprise platforms at IDC, in a statement accompanying the ships and bucks of the server biz.

“In fact, every geographic region except Asia/Pacific experienced revenue contraction in the quarter. It is clear that challenging market conditions are increasing the competitive dynamics for server market share globally, particularly since compute represents a critical element of larger IT transformations that continue to reshape broader enterprise IT market opportunities.”

Gartner pretty much said the same thing in its Q1 server report on Tuesday, although it dices and slices the market slightly differently so you need to have a look-see at both sets of numbers to get a three dimensional view of the server terrain.

IDC concurs with Gartner: The server biz was challenged in the first quarter

By IDC’s numbers, as HP’s server business contracted by 14.8 per cent, nearly twice as fast as the market at large, but it was still able to retain its top spot ahead of Big Blue, which shrank by 13.4 per cent. Both HP and IBM are having issues in their Unix systems businesses, as are Oracle and Fujitsu, and Dell, Cisco Systems, and a bunch of original design manufacturers (ODMs) who build boxes directly for hyperscale data centers are all taking share from them in the x86 racket.

As it is, IDC says Cisco, which grew its sales of blade and rack servers (which only use Intel Xeon processors) by 34.9 per cent to $450m, putting it in a statistical tie with Fujitsu and Oracle. And, interestingly, Fujitsu generated more revenues from server iron than did Oracle, something that Gartner also showed in its numbers and something that has never happened since Sun Microsystems got into the Unix server racket nearly three decades ago.

Oracle is still bleeding revenues like crazy, with another 26.2 per cent drop in the first quarter. Big Larry can talk about how Oracle is committed to Sparc T and M processors as much as he wants, but the sliding has not stopped for two years now and it will not be long before Cisco passes by Oracle.

IDC likes to take a stab at figuring out what the primary operating system is on each box that gets sold, just to give a platform flavor to its numbers. (Yes, we all know machines ship barebones in a lot of cases, but you can put some operating system on it, even a freebie Linux or Unix, and put it in an OS bucket for counting.)

By IDCs guesstimations, the Windows server space is still the biggest part of the server market (as Unix was 15 years ago), accounting for 52.2 per cent of all sales and driving $5.7bn in iron. But Windows declined 4.4 per cent in the quarter just the same.

That is better than dropping at market rate, mind you, but machines that would ultimately run Linux accounted for $2.5bn in sales, up 3.4 per cent. It is hard to believe that Linux will ever catch Windows, but we’ll see how this whole cloud thing plays out in a decade or so.

“Unix systems are the new mainframe” is the nicest way to say it, with Unix-based revenues declining a stunning 35.9 per cent to $1.4bn in the quarter. Yes, IBM, HP, Oracle, and Fujitsu are all smack-dab in the middle of processor transitions, but business had better pick up in the coming quarters or a slew of people are going to lose their jobs.

IBM’s System z mainframes, which missed their sales targets in Big Blue’s first quarter, still had a 7 per cent revenue bump to $800m in the quarter, bucking the overall trend but not as much as IBM needed it to.

By form factor, blade and rack servers both took hits in the quarter as enterprise customers hesitated in their spending, while the density optimized machines preferred by hyperscale data center operators, who either run online services or provide raw and virtualized server capacity to customers on their clouds, continued to spend more money on machines.

This could be workloads shifting, or this could just be the effect of all the consumer stuff we now do on the cloud (that is also sometimes used by businesses, admittedly).

IDC thinks that 235,836 blade servers shipped in the quarter, down 18 points, but ASPs were on the rise thanks to virtualization in the data center and the upsurge of Cisco Systems, which can command a premium for its product because of the integrated networking. (Yes, El Reg is aware of the counter-intuitive nature of that in the serve biz. But some folks just love Cisco.) So blade revenues only shrank by 6 points to $1.87bn.

Vendors shipped 1.05 million rack servers in Q1, down 4.9 percent year-on-year, and prices slipped a bit and so revenues across all vendors were down 5.8 per cent to $5.8bn.

Tower machines lost air on ASPs due to competition (Dell is being very aggressive at the low-end, and HP is really feeling it), and shipments were also down as SMBs are skittish about spending money in many of the regions of the globe, and thus overall revenues for tower machines dropped by 23.8 per cent to $2.45bn.

Those density optimized machines were the bright spot in the market, and very bright indeed, with shipments up 21.7 per cent to 211,172 machines and revenues up 51.4 per cent to $736m thanks to rising ASPs. Dell, by the way, is said by IDC to have a 47.3 per cent share of this density optimized biz.

Prevent-IT neemt NetEyes over

23 mei 2013
By

Ict-dienstverleners Prevent-IT uit Nieuwegein en NetEyes uit Beers gaan per 1 juni samen verder. Hierdoor ontstaat er een organisatie van ruim twintig medewerkers. De naam NetEyes zal blijven voortbestaan. Beide partijen zijn van mening dat de kwaliteit van de medewerkers en de dienstverlening door de overname het best gewaarborgd wordt. NetEyes heeft vooral ervaring met systeem- en netwerkbeheer op de gemeentemarkt met specialisaties die op het gebied liggen van AS/400, SAP, Novell, Linux en Windows.


Bij het in 2001 opgerichte NetEyes werken circa tien medewerkers. Zij blijven in dienst van het nieuwe fusiebedrijf, met uitzondering van de oprichters Stephan Muskens en Frans Wulms.

Prevent-IT heeft zich in zijn twaalfjarig bestaan ontwikkeld tot een brede netwerkdienstverlener, het bedrijf heeft een groot aantal certificeringen behaald, ondermeer voor virtualisatie. Met de overname van NetEyes krijgt Prevent-IT er meer beheerders bij en kan het zich versterken op de overnamemarkt.

Bron: Computable Lees het complete artikel hier: http://www.computable.nl/artikel/nieuws/infrastructuur/4733382/2379248/preventit-neemt-neteyes-over.html

CGI en APG, Bruto Netto Uitkeringssysteem (BNU)

21 mei 2013
By

Computable Awards 2013

Genomineerd voor ICT-project van het jaar in het bedrijfsleven

Tijdens een fixed price, fixed date-project is APGs Bruto Netto Uitkeringssysteem (BNU) in zeven maanden succesvol gemigreerd van een uitbesteed mainframe naar een Unix-omgeving binnen APG. Het betreft het vroegere excassosysteem, dat voor zeven opdrachtgevers met 820.000 deelnemers, maandelijks de pensioenbetalingen verzorgt plus de afdrachten van loonheffingen en zorgbijdragen.


Hiermee is het nieuwe systeem er verantwoordelijk voor dat achthonderd miljoen euro per maand op de juiste rekeningen terecht komt.

De initile doelstelling van het project was om de exploitatiekosten aanzienlijk te verlagen, waarbij de kwaliteit minimaal hetzelfde zou blijven. De resultaten zijn dat APG een besparing van ruim n miljoen euro per jaar realiseert plus de mogelijkheid om het aantal opdrachtgevers uit te breiden zonder dat daar extra kosten mee gemoeid zijn. Volgens de jury een prestatie van formaat.

Terug naar het overzicht met genomineerden

Bron: Computable Lees het complete artikel hier: http://www.computable.nl/artikel/computable_awards/4731690/1853296/bruto-netto-uitkeringssysteem-bnu-apg.html

Bruto Netto Uitkeringssysteem (BNU), APG

21 mei 2013
By

Computable Awards 2013

Genomineerd voor ICT-project van het jaar in het bedrijfsleven

Tijdens een fixed price, fixed date-project is APGs Bruto Netto Uitkeringssysteem (BNU) in zeven maanden succesvol gemigreerd van een uitbesteed mainframe naar een Unix-omgeving binnen APG. Het betreft het vroegere excassosysteem, dat voor zeven opdrachtgevers met 820.000 deelnemers, maandelijks de pensioenbetalingen verzorgt plus de afdrachten van loonheffingen en zorgbijdragen.


Hiermee is het nieuwe systeem er verantwoordelijk voor dat achthonderd miljoen euro per maand op de juiste rekeningen terecht komt.

De initile doelstelling van het project was om de exploitatiekosten aanzienlijk te verlagen, waarbij de kwaliteit minimaal hetzelfde zou blijven. De resultaten zijn dat APG een besparing van ruim n miljoen euro per jaar realiseert plus de mogelijkheid om het aantal opdrachtgevers uit te breiden zonder dat daar extra kosten mee gemoeid zijn. Volgens de jury een prestatie van formaat.

Terug naar het overzicht met genomineerden

Bron: Computable Lees het complete artikel hier: http://www.computable.nl/artikel/computable_awards/4731690/1853296/bruto-netto-uitkeringssysteem-bnu-apg.html

Google’s Native Code browser tech goes cross-platform

18 mei 2013
By

Google I/O At its annual I/O conference in San Francisco this week, Google unveiled a new version of its Native Client technology that allows developers to deploy binary code for web applications in an architecture-independent way.

With the original version of Native Client (NaCl), developers could write modules in C or C++ and compile them into binary packages to be executed inside the browser at near-native speed. The initial release only supported 32-bit and 64-bit Intelx86 architectures, but Google added support for ARM in January.


The revamped version of the tech, dubbed Portable Native Client PNaCl for short, pronounced “pinnacle” still allows developers to target all three architectures, but now they can do so with a single executable package instead of compiling a separate packagefor each processor type.

It works by essentially breaking the process of compiling PNaCl modules into two steps. The PNaCl toolchain really only compiles the C/C++ code halfway, outputting LLVM compiler bitcode instead of native machine code. That bitcode is then packaged into a .pexe file for deployment.

The client browser handles the last mile itself, translating the bitcode of any .pexes it receives into whichever machine code format is appropriate for the platform on which it’s running. That extra compilation step does add a bit of startup overhead, but not too much, and Google has been working to speed up the process even more.

PNaCl does have a few limitations, however. Unlike the earlier NaCl technology, PNaCl does not support linking with glibc, the standard C library found on many Unix-like systems. Instead, developers must statically link their PNaCl modules with the more lightweight Newlib library.

The new version of PNaCl, released on Wednesday, also introduces a new format for .pexe files. Modules created with older, experimental versions of the

Bron: The Register Lees het complete artikel hier: http://go.theregister.com/i/cfa/http://www.theregister.co.uk/2013/05/18/google_portable_native_client/

Ten years on: How did that cloud strategy pan out?

15 mei 2013
By

So the CEO is hearing all about clouds now and the financial director is looking at his pile of beans and as usual wants you to do more with less. And both think it is time for you to build or buy a cloud. Where do you start?

The answer is by being brutally honest with yourself and your bosses about everything around you.


A service provider building a greenfield cloud to peddle infrastructure or platform cloud services to augment your carrier and hosting services has it easy. It is simply a matter of examining what type of cloud it wants to supply to customers.

It picks a cloud controller fabric VMware vCloud, the open source OpenStack or CloudStack, or maybe Windows Server 2012 and Hyper-V with System Center. This cloud doesn’t have to integrate with anything but the providers billing systems: it just has to create a self-service portal for customers and a more sophisticated management console for the providers own admins.

Not so for you. You are sitting there with mission-critical systems physical boxes running siloed workloads or at best virtualized machines that have a few workloads sharing capacity atop a hypervisor.

A fine mess

You probably have a mix of Risc/Unix boxes and maybe some proprietary mid-range and mainframe systems running legacy code.

You have Windows systems running Exchange Servers for email and groupware and any number of SQL Server databases and home-grown apps and third-party apps, and probably Linux systems running other infrastructure workloads such as data warehouses or analytics and maybe Java applications.

Exactly what the mess consists of hardly matters. You have a mix of apps and platforms and developers and admins with their own set of preferences and prejudices. And now the top brass wants you to turn this hodge-podge of hardware and software into a

Bron: The Register Lees het complete artikel hier: http://go.theregister.com/i/cfa/http://www.theregister.co.uk/2013/05/15/cloud_architecture_strategy_avoiding_vendor_lock_in/

Big data opvangen met open hybride cloud

15 mei 2013
By

Wil je als bedrijf mee spelen in de wereld van de big data, dan is het van belang om enkele belangrijke beslissingen te nemen en noodzakelijke investeringen te doen. Een open hybride cloud is volgens mij de manier om op een duurzame en betaalbare manier te investeren in een door big data ondersteunde bedrijfsvoering.


Wat is nu eigenlijk de meerwaarde van big data? Dat is een vraag die steeds meer bedrijven zichzelf stellen. Een ding is zeker: elke organisatie die in zijn bedrijfsvoering veel informatie verzamelt en denkt hier meerwaarde uit te kunnen destilleren, zou nu of op termijn moeten investeren in een systeem dat de grootschalige opslag en analyse van data mogelijk maakt. Daar is een simpele reden voor: traditionele storage-oplossingen voldoen meestal niet aan de noodzakelijke eisen om big data te ondersteunen. Een open hybride cloud is een open source platform met maximale vrijheid en flexibiliteit dat wel eens ideaal zou kunnen zijn om je als organisatie voor te bereiden op big data.

Datarevolutie en OpenStack

Onderzoeksbureau IDC voorspelt in een onderzoek uit december 2012 dat investeringen van bedrijven in big data zullen groeien van 4,5 miljard in 2010 naar 23,8 miljard in 2016. Een van de redenen hiervoor is dat er wereldwijd steeds meer data worden opgeslagen. Dit is duidelijk een trend die het gevolg is van de vraag naar big data-technologie om al deze data op te slaan en er waarde aan te onttrekken. Dit kan op vele manieren, bijvoorbeeld door middel van technieken voor het indexeren van ongestructureerde informatie of het toepassen van statistische analyses op gegevens. Met big data-technologie kunnen bedrijven de data die zij gebruiken inzetten om hun dienstverlening, bedrijfsprocessen en producten te verbeteren.

Open source software wordt veel gebruikt voor het

Bron: Computable Lees het complete artikel hier: http://www.computable.nl/artikel/opinie/storage/4728380/1277017/big-data-opvangen-met-open-hybride-cloud.html

IBM to push Linux apps on Power iron in China, then elsewhere

15 mei 2013
By

IBM is opening a Power Systems Linux Center in Beijing, China, in the hopes of getting more local ISVs interested in its Power Systems iron and luring them away from x86-based systems. With the Power Systems business taking it on the chin in IBM’s first quarter revenues fell 32 per cent compared to a year ago you can bet that Big Blue is trying to light a fire under its Linux-on-Power efforts.

That’s a tough row to hoe, and one that IBM has been engaged in with varying degrees of intensity over the past fourteen years since former CEO Sam Palmisano caught the Linux bug back in 1999 just before he took over running the company.

Palmisano’s replacement as CEO, Ginni Rometty, is now at the helm, and the company’s Systems and Technology Group is bleeding red ink as mainframe sales stall and Power Systems revenues continue to wane with the Unix market.

One response, explained IBM CFO Mark Loughridge on the call going over IBM’s Q1 numbers, is to make Linux more appealing on Power iron, and the establishment of a Linux center in Beijing is one of the efforts Loughridge was referring to.

IBM has been selling Linux-only Power machines, known as PowerLinux boxes, since last April, which were the same iron as regular Power Systems machines but with the prices for processors cut a bit and memory and disk cut a lot to try to close the price gap between Power machines and Xeon servers running Linux.

At the time, IBM reckoned with its Power7 machines as it did when the entry and midrange rack servers were refreshed in February using its eight-core Power7+ processors that the Power-based machines had a considerable performance advantage, and therefore the bang for the buck for

Bron: The Register Lees het complete artikel hier: http://go.theregister.com/i/cfa/http://www.theregister.co.uk/2013/05/14/ibm_power_linux_china_center/