Posts Tagged ‘service level agreements’

Union Solutions takes secure managed hosting downstream to SMBs

Friday, August 31st, 2012

Datacentre formation dilettante Union Solutions has launched an enterprise-class managed hosting use for SMBs.

Its HostPoint charity aims to yield a secure managed hosting sourroundings for business-critical systems and applications.

The use is delivered from Union Solutions’ datacentres to despotic Service Level Agreements, that a association claims provides guaranteed performance, trustworthiness and scalability for particular SMB needs.

Speaking to Cloud Pro, Jason Rabbets, Union Solutions’ co-founder and handling director, said: “SMBs are underneath a lot of vigour currently, coping with a extended retrogression nonetheless being relied on to be a engine of recovery.  This means they need to `stick to their knitting` and concentration on a competitiveness of their products and services and not be dreaming by handling their IT infrastructure.”

With HostPoint, Union Solutions promises 99.9 percent availability, giveaway enterprise-class daily backups, giveaway backup riposte to a second Union Solutions information centre, customary disaster liberation and ISO 27001 and PCI DSS agreeable solutions.

“By rising Hostpoint, Union Solutions is ensuring…SMBs have entrance to an affordable hosting use finish with a infrastructure, support and government formerly customarily accessible for enterprises, complete with giveaway offsite backup, a requirement customarily charged for by other cloud providers,” Rabbets added.

SMBs also have an choice to use an ‘application on-demand’ charging model, permitting any organization to squeeze hosting resources on an as indispensable basis. “Because they are profitable for a use only like a utility, they do not have to foresee hosting mandate adult to 12 months in advance, another cause that will assistance them not only tarry though thrive,” he concluded.

Article source: http://www.cloudpro.co.uk/iaas/cloud-hosting/4463/union-solutions-takes-secure-managed-hosting-downstream-smbs

FIT4Green Project Claims Substantial Datacenter Power Reduction

Tuesday, July 3rd, 2012

July 3 – FIT4Green –project strong in anticipating new solutions for saving appetite in information centres. The plan designed and implemented an energy-aware plug-in on tip of a stream information centres’ government collection to harmonise a allocation of ICT resources and branch off new equipment. Project achieved a goal: 20 % approach ICT apparatus appetite assets but compromising correspondence with Service Level Agreements (SLA) and Quality of Service (QoS) metrics. The achieved assets in CO2emissions were on a same scale as in energy. The approach appetite assets in a ICT apparatus satisfy also conspicuous additional assets due to a reduced needs for cooling, for example.

FIT4Green plug-in is designed to be germane to any information centre type. The plug-in was certified in 3 deputy information centres: service/enterprise portal during ENI, supercomputing information centre during Jülich Supercomputing Centre with a federated site during VTT Technical Research Centre of Finland, and cloud computing height during HP. VTT’s work in a plan strong on a optimizations in a supercomputing scenario. The aim of 20% was reached in any exam bed, and in some cases a assets were even adult to 50%. The comparison indicate for all a assets was a same complement but any appetite optimizations.

All a 16 open deliverables of a plan are openly accessible on a plan web site at http://www.fit4green.eu. The plug-in formula has also been expelled as open source software.

FIT4Green was concurrent by GFI Informática with HP Italy Innovation Centre as a technological leader. Other partners besides VTT were University of Passau, Jülich Supercomputing Centre, Imperial College London, University of Mannheim, Create-Net, Eni S.p.A., and Almende BV.

—–

Source: FIT4Green

Article source: http://www.hpcwire.com/hpcwire/2012-07-03/fit4green_project_claims_substantial_datacenter_power_reduction.html

The Best SLA Ever

Monday, March 5th, 2012

This post is partial of a ReadWriteCloud channel, that is dedicated to covering virtualization and cloud computing. The channel is sponsored by Intel and VMware. Read a box investigate about how Intel Xeon processors and VMware helped virtualize 12 business vicious database applications.

singlehop-logo.jpgYou no doubt are rather asocial about use turn agreements (SLAs), those little-reviewed papers that guarantee a turn of use from your hosting provider. Little-read that is, until something goes awry. Enter SingleHop, a Chicago-based provider that is perplexing to make a name for itself by indeed delivering a solid “Bill of Rights” for customers and earnest to compensate when they don’t accommodate their SLA. It is an engaging idea.

There are lots of other providers that do offer payouts for missed use milestones, of course. But we favourite a approach they are creation it easy to see their metrics and what we will get when they are off. Take a demeanour during a screenshot below:
sla2.jpg

For example, if we have to wait some-more than 3 hours to spin adult a dedicated server, we can take 10% credit on your subsequent bill. (Like that piggybank icon? A good touch.) Or wait some-more than an hour to a initial responder to your difficulty ticket? Take 5% off. One of a facilities singular to their Bill of Rights is a news card-like functionality that allows SingleHop’s business to see how good their comment is being handled. The minute news shows use annals as they review to a company’s guidelines.

SingleHop has dual opposite SLA tiers: a unchanging SLA (where we took a screenshot) and an extended SLA that has faster guarantees and aloft credits. All support requests from SingleHop are rubbed by American employees, and an tangible worker and not partial of a call center. Now that is bringing a new turn of use too.

Posted in

and tagged with

You no doubt are rather asocial about use turn agreements (SLAs), those little-reviewed papers that guarantee a turn of use from your hosting provider. Little-read that is, until something goes awry. Enter SingleHop, a Chicago-based provider that is perplexing to make a name for itself by indeed delivering a solid “Bill of Rights” for customers and earnest to compensate when they don’t accommodate their SLA. It is an engaging idea.
Please capacitate JavaScript to perspective a comments powered by Disqus.

Article source: http://www.readwriteweb.com/cloud/2012/03/the-best-sla-ever.php

Symantec fends off VMware for information centre throne

Wednesday, October 5th, 2011

Symantec Vision Symantec is favourable a Storage and Availability Management program (SAMG) for a virtualised cloud universe with a recover of chronicle 6.

The association is pulling a multi-vendor certification and aims to accelerate a software’s purpose in security, accessibility and storage government in information centres. However, it faces dual challenges: incoming foe from VMware and expanding storage array controller functionality. With this latest SAMG revision, Symantec is observant we can emanate a non-trivial cloud use regulating a infrastructure we already have.

The products updated in a SAMG 6 launch are:

  • Veritas Storage Foundation 6.0
  • Veritas Cluster Server 6.0
  • Veritas Operations Manager 4.1
  • Symantec ApplicationHA 6.0
  • Veritas Dynamic Multi-Pathing 6.0
  • Veritas Cluster File System 6.0
  • Veritas Replicator 6.0
  • Symantec VirtualStore 6.0
  • Veritas Storage Foundation for Windows 6.0
  • Veritas Storage Foundation for Oracle RAC 6.0
  • Veritas Storage Foundation for Sybase ACE 6.0

Veritas Operations Manager 4.1 breaks a 6.0 nomenclature; that’s odd, maybe there’s some-more to come there.

The upgraded program should cope a lot improved with multi-tiered applications that yield a business service, and safeguard a insurance and accessibility of practical business services (VBS). This should be probable opposite opposite handling systems and virtualisation regimes, with a VBS encapsulating a multi-tier business app set adult as a singular entity:

Administrators can design to harmonise a start, stop, non-disruptive testing, and failover of whole multi-tier applications with a singular click.

Symantec says that, handling in this way, it can assistance a business accommodate cloud use turn agreements (SLAs) as they muster private, hybrid and open clouds. With Veritas Operations Manager and Veritas Cluster Server installations, business are betrothed a perspective of accessibility and disaster liberation standing opposite a information centre from a singular monitor.

In other difference Symantec is staking a explain to be a information centre’s operations government apparatus and not ceding that purpose to pretender VMware.

The Veritas Storage Foundation (VSF) apartment is dictated to work opposite a multi-vendor environment, to grow and cringe provisioned storage in genuine time, optimise a use of array capacity, and capacitate multi-vendor storage choice. VSF has support for “thin reclamation, storage tiering, energetic resizing, and intelligent migration”.

Storage Foundation 6.0 is designed to support deduplication and focus during a record complement layer. We don’t know if it will pierce to retard storage as well. Symantec says: “The Veritas Storage Foundation apartment includes Veritas Dynamic Multi-Pathing, Veritas Cluster File System, and Symantec VirtualStore, with support for all vital storage arrays and handling complement platforms.”

In a approach Storage Foundation is an enabler of commoditised storage and Symantec is pulling a summary that what enterpriser array controller program does it can do with typical arrays.

Symantec is staking a explain to be a information centre’s operations government apparatus and not ceding that purpose to pretender VMware.

The v6.0 program introduces a new pay-for-what-you-use pricing model.

The gymnasium for a keynote event announcing this things was filled with happy, smiling people, and because not? They’d paid their assemblage price and got news of a comprehensive, opposite a house program refurbish assisting them lift on doing what they are already doing: facing a summons songs of single-source vendors like VMware, EMC and NetApp.

There was good news on a confidence front too with a raft of announcements. One prominence was a phenomenon of focus repute management. This involves a database of some 3.1 billion applications. If we pointer adult for this and your complement tries to run an app that’s not in a database afterwards it’s suggested to we that we don’t run it.

We’re meditative of this as a quasi-industry-wide whitelist facility. CEO Enrique Salem was flattering gratified with a idea. So was his audience, and Symantec’s repute went adult a nick as a result. ®

Article source: http://www.channelregister.co.uk/2011/10/05/symantec_samg_6_launch/

vSphere 5′s Storage DRS and storage form duty broach control over storage resources

Saturday, September 24th, 2011

The recover of VMware Inc.’s vSphere 5
brings many sparkling new facilities and enhancements to a virtualization platform, generally when
it comes to storage. Two of a biggest new facilities in that area are Storage Distributed Resource
Scheduler (DRS) and Profile-Driven Storage, that yield some much-needed control over storage
resources.

In prior versions of vSphere, Distributed Resource Scheduler
balanced VM workloads formed on CPU and memory apparatus utilization. Storage
DRS
extends this capability to storage, enabling intelligent VM initial chain and load
balancing formed on storage I/O and ability conditions within a cluster. Profile-Driven Storage,
for a part, ensures that VMs are placed on storage tiers formed on service-level
agreements
(SLAs), availability, opening and capabilities of a underlying storage
platform. In this tip, we’ll examine


both Storage DRS and a storage form functionality in
detail.

Storage DRS

Similar to a normal DRS feature, Storage DRS uses a new form of cluster called a data
store cluster, that is a collection of information stores that are many-sided into a singular territory of
consumption. By determining all of a storage resources, Storage DRS allows intelligent placement
of VMs that are powered on, as good as a changeable of workloads from one storage apparatus to
another when indispensable to safeguard best opening and equivocate I/O bottlenecks. What this means in
simpler terms is that, identical to vMotion’s
movement of VMs from horde to host, VMs can now be changed from information store to information store as well; the
decision to pierce a VM from one information store to another is done by Storage DRS, that tells Storage vMotion to
make a move.

Data store clusters are combined by going to a Datastores and Datastore Clusters intent in the
Inventory territory of a vSphere
Client
and right-clicking on a information core and selecting a New Datastore Cluster option. This
will launch a sorceress that allows we to configure a cluster and conclude automation settings.

At a initial screen, name a cluster and select possibly to capacitate Storage DRS or not. Then
choose an automation turn for Storage DRS—either primer mode, where it creates recommendations but
does not act on them, or full programmed mode, where VM hoop files are automatically moved. Then
define a runtime manners for Storage DRS, that mention how it operates. You can select to include
I/O metrics as partial of Storage DRS recommendations or not. If we do not embody them, only
utilized space is factored in. You can also conclude a thresholds for both employed space and I/O
latency. Utilized space can be set from 50% to 100%; this environment dictates a smallest turn of
consumed space that is a threshold for action. The default is 80%; with that setting, if a data
store has some-more than 20% giveaway space, no movement is taken. I/O latency can be set from 5 milliseconds
(msec) to 50 msec; this dictates a smallest I/O throughput indispensable to take action. The default is
15 msec, that means during slightest 15 msec of I/O latency contingency be occurring before movement is taken.

There are also modernized settings where we can conclude a disproportion that contingency start between the
source and finish information store for movement to occur. For example, if a threshold is set to 5
and if a space used on a source information store is 82% and on a finish information store it’s 79%,
Storage DRS will not make emigration recommendations from a source information store to a destination
data store. You can also conclude how mostly vCenter Server
re-evaluates I/O workloads. The default is 8 hours, that is not really frequent. The
re-evaluation can be set in minutes, hours or days. Finally, there is an I/O imbalance environment that
performs migrations usually when a I/O bucket imbalance exceeds a threshold.

Once we conclude your Storage DRS runtime settings, we afterwards select that hosts to embody in the
cluster as good as that information stores, and afterwards we are prepared to start regulating Storage DRS.

One vicious underline for determining Storage DRS function is anti-affinity rules. These are
used to safeguard that specific VMs or specific practical disks do not finish adult on a same information store
together. There are certain situations when we wish to keep VMs apart—for example, if we have two
heavy I/O database servers regulating on apart VMs, we might not wish both those workloads thrashing
the same information store. Another is for error toleration reasons: If we have dual VMs of that during least
one contingency sojourn running, by gripping them on apart information stores, we can safeguard that if one data
store fails, one of a VMs will continue to run.

While Storage DRS is a good feature, we should be clever with it and make certain we do not set
it too aggressively. A Storage vMotion charge is really I/O-intensive. If too many start during a same
time, opening can suffer. This emanate can be during slightest rather mitigated by a use of a
storage array that provides duplicate offloads for Storage vMotion so horde resources are not consumed
when they occur. And it’s best to initial use Storage DRS in primer mode before we try automated
mode, to guard recommendations and get a feel for what will occur.

Profile-Driven Storage

Profile-Driven Storage enables we to safeguard that VMs are stored on inclination that have specific
characteristics, such as certain capacity, availability, opening and excess levels. For
example, if we have a vicious VM that requires high storage performance, we can make certain that
it uses usually Fibre Channel information stores instead of iSCSI information stores.

There are dual components to this feature: storage capabilities and VM form documents. Storage
capabilities can be possibly system-defined or user-defined. System-defined capabilities are
automatically populated by storage arrays that support a new vStorage APIs for Storage Awareness
(VASA). Using VASA, a storage array can tell vCenter Server a capabilities, such as which
features it supports, opening characteristics, redundancy, capacity, etc. vCenter Server
assigns a system-defined storage capability to any information store that we emanate from that storage
system. If a storage array does not support VASA, we can still manually conclude capabilities for a
storage device. You can afterwards associate user-defined capabilities to a information store. You can associate
a user-defined capability with a information store that already has a system-defined capability. But a
data store can have usually one system-defined and usually one user-defined capability during a time.Once you
have storage capabilities defined, we afterwards emanate VM storage profiles from among all a storage
capabilities that are available. These profiles are used during provisioning, cloning and Storage
vMotion to safeguard that usually those information stores or information store clusters that are agreeable with the
virtual appurtenance storage form are done available.

Once we have profiles created, we can afterwards allot them to a VM by right-clicking on a VM and
selecting a VM Storage Profile choice and selecting Manage Profiles. You can also revise the
settings of a VM, and on a Profiles add-on we can allot profiles to a VM and a practical disks
(.vmdk). A VM’s application files (.vmx, .vmsd, .nvram, .log, etc.) and a practical disks can have
separate VM storage profiles. Once profiles are enabled and reserved to VMs, we can afterwards check
whether a VM and a practical disks are regulating information stores that accommodate a correspondence that is set in
the VM storage profile. In a VM Storage Profiles territory of a vSphere Client, we can
configure, capacitate and check a correspondence standing of VMs. For noncompliant VMs, we can select to
migrate them to a information store that will prove compliance.

Eric Siebert is a VMware consultant and author of dual books on virtualization.

This was initial published in Sep 2011

Article source: http://www.pheedcontent.com/click.phdo?i=d7f8bffd24255c50081fa64c41ce9c56

Cloud word sees misty future

Tuesday, September 20th, 2011

Service-level agreements (SLAs) per to cloud-computing services are now extensive adequate for any use disruption, records an attention analyst. One word vendor, however, says that cloud-specific word services can be complementary, and offer many marketplace potential.

Chris Morris, investigate executive of cloud technologies and services during IDC Asia Pacific, questioned a need for cloud word packages, observant that given a resources are owned by a cloud use provider, it is adult to this provider to recompense any detriment of income to business in a eventuality of a use interruption.

The researcher told ZDNet Asia that opposite IT applications rely, to varying extents, on outwardly sourced cloud computing services. If a cloud use is elementary and clearly defined, such as a simple infrastructure as a use for non-critical focus hosting, Morris pronounced that penalties settled in customary SLAs should be sufficient, and do not need additional insurance.

For some-more formidable and business-critical deployments, in that a use is delivered by mixed providers, such as telcos and server providers, he pronounced that a responsibility here is on a focus owners to entirely know a impact on their business if there is a detriment to a service.

“If it is that vicious to a delay of income for a business user, afterwards a risk criticism will brand this, and a governance cabinet will insist on business word to be carried by [the cloud vendor],” combined a IDC analyst.

Difficulties in risk assessment

Asked about a stipulations of putting together an word package for cloud-based services, Morris reiterated that many problems arise from a inlet of cloud services, and how mixed vendors are endangered in a smoothness process.

He remarkable that unless a singular businessman agrees to take on a lead provider purpose in such contracts, risk criticism of a cloud use will be “difficult”, and a word reward is doubtful to be competitive. Conversely, if a singular businessman can control a whole use smoothness supply chain, it would afterwards be some-more unsentimental for a third-party word services provider to yield word for it, creation a word package some-more appealing to intensity customers, he said.

Morris remarkable that in many cases, a chastisement payments charity by cloud use providers did not simulate a intensity for mislaid income incurred by influenced customers. In such cases, where a business risk was large adequate to aver coverage, companies should afterwards demeanour into purchasing a applicable policies.

That said, he combined that a “downside” to this choice was that cloud services were corroborated by usually a brief lane record for a smoothness of vicious business services. In a deficiency of explanation of delivery, a premiums for insurances to vicious services competence be so costly as to be unaffordable, a researcher surmised.

Insuring data, not service

However, Meghan Hannes, handling executive during cloud word provider CloudInsure, that was started in Aug 2010, pronounced that a association is holding a opposite proceed to cloud word than what Morris described.

Hannes told ZDNet Asia in a phone talk that rather than yield coverage for a cloud use per se, CloudInsure is conceptualizing a policies for “cyber liability”, that includes aspects such as regulatory coverage, information remoteness and security. In other words, it assesses a risks associated to a form of information that a association is exposing with cloud-based services, she explained.

Additionally, she pronounced that a company’s products — approaching to be accessible in a fourth entertain of this year or a initial entertain of 2012 — would “not interfere” with existent SLAs. Rather, a word would support these agreements and “starts where a SLAs finish off”.

According to CloudInsure’s website, it now lists 6 cloud providers in a initial pool of cloud services it rates. These are Amazon Web Services (AWS), NetSuite, Oracle, Salesforce.com, Intuit and SuccessFactors. How CloudInsure does a rating, though, stays underneath a non-disclosure agreement, Hannes said.

Quizzed on either CloudInsure was a initial to offer such word services, she replied that it was. Elaborating, she concurred that there were already existent cyber guilt services that support to a detriment of information on a internet or corporate websites. However, a marketplace for such services assumes that a infrastructure is owned by a company, since CloudInsure takes a position that a infrastructure providing a cloud services is not owned by a customers, she explained.

With regards to how a association labelled a word premiums, Hannes suggested that this was still in a “development phase”, though a running element would be “pricing risk according to a value of a information being insured”, that she pronounced reflected “true mercantile loss” in times of use disruption.

She forked out that industries where there were “regulated data”, such as medical and financial services, would be one of a “first movers” in terms of adopting such word policies, as these organisations faced some-more mercantile guilt for a information they placed on cloud services. As such, Hannes pronounced that there was a “tremendous volume of untapped potential”, where cloud word was concerned, and a marketplace could be value anything in a millions to billions range.

One cloud use provider that ZDNet Asia spoke to, per a viability of cloud word products, chose not to criticism on a doubt directly. Instead, Amazon Web Services’ orator Regina Tan said: “We will continue to listen to business on what is many critical to them with courtesy to SLAs, and will always cruise ways in that we can raise a charity for customers.”

Tan combined that cloud computing reinforces some aged concepts of building rarely scalable internet architectures, while introducing some new concepts that change a approach applications are built and deployed. As such, companies looking to precedence a full advantages of a cloud, including a agility and scalability, need to initial know a services, facilities and best practices, she suggested.

In April, AWS suffered a prejudiced disaster to a Elastic Compute Cloud (EC2) during a northern Virginia site, that brought down important sites such as Quora and Reddit. Its Dublin, Ireland, EC2 site also gifted downtime final month, following an “explosion that caused a transitory electric deviation” and disabling some of a energy synchronisation systems, according to a report by record site Data Center Knowledge.

Via ZDNet Asia

Article source: http://www.zdnet.com.au/cloud-insurance-sees-hazy-future-339322676.htm

IBM has the conduct in a cloud in SA

Saturday, September 17th, 2011

The fast expansion of broadband services as a outcome of fibre-optic connectivity is pushing a company’s concentration on sub-Saharan Africa, pronounced Werner Lindemann, vice-president of tellurian record services for IBM in a region.

Cloud computing allows companies to cut their IT capex by usually profitable for what they use. It creates IT services accessible online instead of on a device. A pay-as-you-use pricing indication allows companies to simply envision their handling expenses.

According to Lindemann, this is a initial event for clients in South Africa to advantage from hybrid private cloud offerings that offer fixed use turn agreements while also adapting to governance issues of information storage.

IT investigate organisation IDC pronounced worldwide income from open IT cloud services exceeded $21.5-billion in 2010 and will strech $72.9-billion in 2015, representing a devalue annual expansion rate of 27.6%.

Article source: http://www.businesslive.co.za/southafrica/sa_companies/2011/09/17/ibm-has-its-head-in-a-cloud-in-sa

IBM SA expands business with new cloud information centre, lab

Wednesday, September 14th, 2011

IBM South Africa announced an enlargement of a cloud computing services for clients in southern Africa, with a further of a new IBM Cloud Data Centre and Cloud Lab, in Johannesburg.

The new additions to a existent Business Continuity and Recovery Service centre (BCRS) will extend IBM’s globally integrated cloud smoothness network of cloud computing centres that offer in over 50 countries around a world, with centres formed in Singapore, Germany, Canada and a US; and 13 tellurian cloud labs.

The new IBM Cloud Data Centre will broach IBM’s SmartCloud enterprise-class services, that embody a extended spectrum of secure managed services, to run different workloads opposite primarily hybrid private cloud smoothness methods.

According to Werner Lindemann, Vice-President for Global Technology Services during IBM Sub-Saharan Africa, this is a initial event for craving clients in South Africa to advantage from hybrid private cloud solutions that offer fixed use turn agreements on a program as a use (SaaS) or focus computing-based model, while also adapting to governance issues of information storage.

According to IDC, worldwide income from open IT cloud services exceeded $21.5 billion in 2010, and will strech $72.9 billion in 2015, representing a devalue annual expansion rate (CAGR) of 27.6%.

The SmartCloud offerings capacitate clients to name pivotal characteristics of a public, private and hybrid cloud to compare effort requirements, from elementary Web infrastructure to formidable business processes, along 5 dimensions, including:

* Security and siege
* Availability and opening
* Technology platforms
* Management support and deployment
* Payment and billing

New offerings will boost patron choice, with a intensity for end-to-end supervision of use smoothness from a server and handling complement to a focus and routine layer.

The Cloud Lab trickery will assistance businesses, supervision and investigate institutions, and institutes of aloft learning, to design, adopt and reap advantages of cloud technologies, as good as assist IBM Business Partners cloud-enable their technologies and benefit a sales and selling skills they need to take advantage of this fast-growing market.

IBM Cloud Computing

IBM has helped thousands of clients adopt cloud models, and manages millions of cloud-based exchange each day. IBM assists clients in areas as different as banking, communications, medical and supervision to build their possess clouds or firmly daub into IBM cloud-based business and infrastructure services. IBM is singular in bringing together pivotal cloud technologies, low routine knowledge, a extended portfolio of cloud solutions, and a network of tellurian smoothness centres. For some-more information about IBM cloud solutions, revisit http://www.ibm.com/smartcloud.

 

Editorial contacts

Text 100 Johannesburg
Boitumelo Mkhondo
(+27) 011 775 5711
This e-mail residence is being stable from spambots, we need JavaScript enabled to perspective it

 



<![CDATA[

]]>

Subscribe NOW! Essential Virtualisation news filtered for we

Enable Javascript in your Browser

Article source: http://www.itweb.co.za/index.php?option=com_content&view=article&id=47185:ibm-sa-expands-business-with-new-cloud-data-centre-lab&catid=279

DevOps: Keep tabs on cloud-based app performance

Friday, September 9th, 2011

Poor opening of open cloud-based applications leads to finish user frustration. SLAs from
PaaS and IaaS providers cover availability, though not altogether response time. DevOps teams can monitor
cloud computing
apps for opening regulating a few opposite tools.

More on apps in a cloud:

Cloud
shakes adult focus smoothness in a enterprise

How
providers impact cloud focus migration

Mapping
applications to a hybrid cloud

Nearly 350 IT pros via North America estimated
an normal annual income detriment of $985,260
from opening problems with cloud-based
applications, reported a Compuware survey. Respondents in a EU estimated a $777,000 detriment (in U.S.
dollars). Concerns about app
performance
caused 58% of North American and 57% of EU-based respondents to check adoption of
cloud-based applications.

The news also found that 94% of North American respondents and 84% of EU respondents think
cloud focus service-level
agreements (SLAs)
are formed on a tangible end-user experience, not only use provider
availability metrics. This naivety expected won’t remonstrate open or private cloud providers to offer
potentially dear end-to-end SLAs. Therefore, it is DevOps’ pursuit to instrument cloud-based apps
with blunder logging, analytics and evidence code.

So what’s a best approach to guard opening of your data-intensive applications? Free or
low-cost uptime and response time reports from Pingdom.com,
Mon.itor.us
and other site monitoring providers endorse that applications accommodate SLAs and end-to-end site
performance.

Firms like LoadStorm
and Soasta sell customized cloud application
load testing. Compuware’s CloudSleuth site provides a giveaway monthly Global
Performance Ranking of end-to-end response times of vital open Infrastructure
as a Service (IaaS)
and Platform
as a Service (PaaS)
providers.

Figure 1 shows a Windows Azure representation plan that demonstrates paging as good as create, read,
update and undo (CRUD) operations on an Azure table. Attempting this for OakLeaf’s Azure Table Services Sample Project would
result in disastrous values for formula execution.


FIGURE 1. Windows Azure representation plan arrangement paging and CRUD operations.

The Time content box during a bottom of a window indicates a final action’s formula execution time,
which was 28 ms for a new page. By clearing a Batch Updates check box, you’ll see a dramatic
increase in execution time for particular CRUD operations on a 91 patron records.

Relying on reports isn’t enough; organizations also contingency acquire or rise on-premises
diagnostic government collection to download and investigate opening logs, as good as to broach alarms
and graphical reports. IaaS providers like Amazon Web Services (AWS) combine on building their hardware with native
metrics; PaaS products like Windows Azure and SQL Azure yield deeper, some-more customized insight
into a focus and a code.

How AWS CloudWatch keeps an eye on performance 
Amazon CloudWatch allows DevOps
teams to automatically guard CPU, information transfer, hoop activity, latency and ask depends for
their Amazon Elastic Compute Cloud (EC2) instances. Basic metrics for EC2 instances, EBS volumes,
SQS queues, SNS topics, Elastic Load Balancers and Amazon RDS database instances start at
five-minute intervals for no additional cost. You can supplement customary metrics and alarms in a AWS
Management Console’s CloudWatch add-on and perspective graphs of metrics by navigating to a Metric page in
the Navigation Pane (Figure 2).


FIGURE 2. A graph for an EC2 instance’s CPU duty metric, many-sided by EC2 Image ID. The
remaining 6 metrics contain a 7 Basic Metrics for EC2 images.

The DevOps group can use AWS’s Auto
Scaling
underline to yield effervescent accessibility by adding or deletion Amazon EC2 instances
dynamically formed on an app’s CloudWatch metrics. AWS combined new notification, regularity and other Auto Scaling features in Jul 2011.

CloudWatch doesn’t yield built-in focus monitoring metrics given AWS is an IaaS
offering that’s OS- and development-platform agnostic. However, developers can procedure applications
to submit API requests in response to app events, such as rubbed or unhandled
errors, duty or procedure execution time, and other app-related metrics. 

Windows Azure opening logging and analytics 
The Windows Azure group has been adding logging, analytic and evidence facilities to a platform
steadily given a PaaS use became accessible in Jan 2010. Because Windows Azure Portal’s
interface doesn’t support adding metrics, analytics or alarms, DevOps teams contingency write formula and
edit pattern files to capacitate diagnostics and logging for Windows Azure discriminate and storage
services. Table 1 shows logs that were accessible for research by DevOps with Windows Azure’s
Diagnostic API in Jun 2010.


TABLE 1. Windows Azure evidence information collection logswith a names of Azure tables
and binary vast objects (blobs) that store a record information enabled by Windows Azure SDK v1.2.

SDK v.1.4 and Visual Studio Tools for Azure v1.4 combined a capability to profile a Windows Azure
app
with Visual Studio 2010 Premium or Ultimate when it runs in Windows Azure’s production
fabric. An early Aug 2011 refurbish to collection and a new Windows Azure Storage Analytics feature enabled logs that snippet executed
requests for storage accounts as good as metrics that yield a outline of ability and request
statistics for binary vast objects (blobs), tables and queues.

An updated sample plan of Aug. 22 generated analytics tables for both list and blob
storage, as good as inner timing information formed on TraceWriter items, though requires a apart app to
read, arrangement and conduct diagnostics. Cerebrata’s Azure Diagnostics Manager reads and displays record and diagnostics
data in tabular or striking format. The firm’s giveaway Windows Azure Storage Configuration Utility lets IT operations teams spin on
storage analytics but essay code. Cerebrata’s Cloud Storage Studio adds list and blob government capabilities
(see Figure 3).


FIGURE 3. A beta chronicle of Cerebrata’s stirring Cloud Storage Studio v2.

The System
Center Monitoring Pack for Windows Azure Applications
provides capabilities identical to
Cerebrata products for System Center Operations Manager (SCOM) 2007 and 2010 beta users.
Autoscaling Windows Azure for effervescent accessibility now requires a third-party solution, such
as Paraleap’s AzureWatch, hand-crafted .NET functions or a stirring Windows Azure Integration Pack for Enterprise Library. Monitoring SQL Azure
databases requires we to use SQL Azure’s 3 categories of dynamic management
views
: database-related views, execution-related views and transaction-related views.

ABOUT THE AUTHOR
Roger Jennings is a data-oriented .NET developer and writer, a principal consultant of OakLeaf
Systems and curator of a OakLeaf Systems blog. He’s also a author of 30+ books on a Windows Azure
Platform, Microsoft handling systems (Windows NT and 2000 Server), databases (SQL Azure, SQL
Server and Access), .NET information access, Web services and InfoPath 2003. His books have some-more than 1.25
million English copies in imitation and have been translated into 20+ languages.



This was initial published in Sep 2011

Article source: http://www.pheedcontent.com/click.phdo?i=95de5e3e73b264a6d6a125748f36c835

Managed services kick “baby datacentres”

Wednesday, September 7th, 2011

<!—->

Uptake of a managed services indication is flourishing locally, according to Peter Macaulay, principal, executive programme during IDC New Zealand.

But a change from in-house operations to a managed services or outsourced resolution is a outrageous change, he says. Traditionally CIOs have had a clever operational focus. Under a new models, they are ceasing to run any apparatus and are instead focusing on environment adult use turn agreements and handling relations with vendors that broach those services for them.

“It’s a opposite indication of doing business and it’s utterly a tough transition to make. The tangible transition is not easy and requires poignant management,” he says. “It’s not a matter of gnawing your fingers and saying: ‘we’re putting all in a cloud’.”

The pivotal to a successful transition is carrying people onboard that know a stream operational indication and nonetheless have a ability to adopt a creation and change in processes that are indispensable to pierce into a cloud-type mode, says Macaulay.

This indication is “quite a vast shift” and it requires a opposite mindset.

“To be fair, many CIOs onslaught with a change in mindset.”

Macaulay sees 3 stairs of relocating to a managed services indication – a initial square is a aged indication with a pack kept and managed in-house; a third partial is where all a pack is outsourced and we are handling relations and SLAs; a center bit of transitioning is enormously formidable and complex.

“Not each CIO can hoop that, and not each CIO can do all 3 of those roles.”

Organisations need to cruise if they unequivocally need to demeanour after their applications themselves, he says. And even if they do, is it improved to have them regulating on servers that are managed by someone else? In many cases, that indication is mostly hugely profitable to a organisation, says Macaulay.

“If we are doing a vital development, we can go to your use provider and say: ‘I’m going to need X series of CPUs for a subsequent 3 months – yield them, please’. At a finish of 3 months we can move them down.”

No longer do we have to have a smoke-stack of servers sitting on stand-by, or worry about progressing and updating them, he says.


Better use of time during Abano

Abano Healthcare Group in Auckland is enjoying only those benefits, according to a CIO, Peter Radich.

The organization is an user of medical and medical services businesses in New Zealand, Australia and Asia, with services in audiology, dental, diagnostics and rehabilitation. The organisation, that has some-more than 200 users locally, uses a “mini-cloud” resolution from Appserv, also formed in Auckland.

“Appserv runs radically all a applications in a datacentre and broach them over a skinny client, regulating Citrix desktop,” says Radich.

Abano Healthcare has been an Appserv patron for some-more than 5 years and has no skeleton of changing things, says Radich.

From an IT indicate of view, a managed services indication is some-more secure, some-more cost-effective and some-more controlled, Radich says. Abano Healthcare gets entrance to a scale and record of a most incomparable organisation.

The association wouldn’t be means to have a same resilience if it had a possess “little baby datacentre”, he says.

“It also means we get entrance to collection and people that we couldn’t means otherwise.”

It’s dear to have network engineers, systems engineers and confidence experts on staff. “Essentially, it’s a ability to lease rarely learned IT staff and not carrying to compensate their full costs,” he says.

Appserv also takes caring of backup and disaster liberation for Abano, that differently had been dear for a association to set adult itself.

The day-to-day IT operation is now one reduction thing to worry about, he says.

“I can now concentration my time on business projects, on how to use IT to make a business better, faster and some-more volatile and reduce a cost overall, rather than spending time on Exchange upgrades or worrying about licensing,” Radich says.

He now spends about 80 percent of his time of new projects. “It’s a improved use of my capability,” he says. “I don’t have to have a vast team. In fact, we don’t unequivocally have a team. Our partners assistance us broach projects.”

“For only gripping a lights on – it’s improved to get somebody who is unequivocally good during that to demeanour after that area. [Appserv] focuses on it 100 percent, all day, each day.”

But we have got to be clever when selecting your managed services provider, says Radich.

Abano had a same outsourcing truth for a businesses in Australia and Asia, and “that whole routine has shown that it’s vicious to get a right distance outsourced partner – somebody who is a identical distance to we so that we are not mislaid in a mix”, he says.

“As a small-to-medium craving we wouldn’t tend to use a vast [outsourcing partner] since we get mislaid in a noise. we like to be means to phone a chairman who owns a business and tell him what we think, or what we need done.”

And we have got to have “a bloody good relationship” with your partner, since we are tied in and it has got to work. As critical is carrying plain SLAs, so both parties know who looks after what and what turn of use we require.

Also, don’t skimp, says Radich. Make certain we compensate a right volume of income for a right service. “There is no such thing as cheap, good IT,” he says.
<!—->

Article source: http://computerworld.co.nz/news.nsf/special/managed-services-beat-baby-datacentres