Performance Report
Cloud storage lives up to the hype
Aug 23rd
Very Interesting Report By Tom Henderson and Brendan Allen, at Network World
In our continuing series of groundbreaking tests of cloud computing services, we take a look at what enterprises can expect if they decide to entrust data to a cloud storage provider. We found that cloud storage lives up to its advance billing in two key areas: cloud storage can be fast and the pay-as-you-go model can be a real cost saver. We also found that security could be an issue for enterprise shops, and the formulas for trying to predict overall costs can be complex.
The services that we tested were Amazon S3, Rackspace’s CloudFiles, Egnyte’s On Demand File Server, Nasuni Cloud Storage, and Nirvanix‘s Storage Delivery Network. Amazon, Rackspace and Nirvanix represent the containerized/object-oriented model. Egnyte embodies the file/folder metaphor, while Nasuni offers a different twist – it’s a front-end that simplifies cloud storage for enterprise customers and connects to other cloud storage vendors on the back end.
Related posts
Top 10 Cloud Computing Load Test and Performance Monitoring Companies
Aug 20th
Top 10 Cloud Computing Load Test and Performance Monitoring Companies
Keynote
Test Perspective is a cost-effective, completely self-service With Test Perspective you can run the most realistic load tests on-demand and . With Test Perspective you can run the most realistic load tests on-demand and receive immediate feedback on modifications you make to your Web site.
Soasta
Is available as an on demand service in the cloud or as a physical or virtual appliance, SOASTA CloudTest’s seamless integration of test design, monitoring, and reporting offers everything you need to test and deliver high quality Web applications and services at an affordable price.
Monitis
With its Universal Cloud Monitoring Framework, Monitis can now sync to other Cloud computing providers very quickly – from Rackspace, GoGrid, 
Softlayer, and more. Monitis’ Universal Cloud Monitoring Framework will automate monitoring in highly dynamic cloud environments, where customers’ servers maybe added and terminated according to the load by management software or manually.
Related posts
End-To-End Performance Study of Cloud Services
Jun 3rd
Here is a good study found at: High Scalability
Cloud computing promises a number of advantages for the deployment of data-intensive applications. Most prominently, these include reducing cost with a pay-as-you-go pricing model and (virtually) unlimited throughput by adding servers if the workload increases. At the Systems Group, ETH Zurich, we did an extensive end-to-end performance study to compare the major cloud offerings regarding their ability to fulfill these promises and their implied cost.
The focus of the work is on transaction processing (i.e., read and update work-loads), rather than analytics workloads. We used the TPC-W, a standardized benchmark simulating a Web-shop, as the baseline for our comparison. The TPC-W defines that users are simulated through emulated browsers (EB) and issue page requests, called web-interactions (WI), against the system. As a major modification to the benchmark, we constantly increase the load from 1 to 9000 simultaneous users to measure the scalability and cost variance of the system. Figure 1 shows an overview of the different combinations of services we tested in the benchmark.
 |
| Figure 1: Systems Under Test |
The main results are shown in Figure 2 and Table 1 – 2 and are surprising in several ways. Most importantly, it seems that all major vendors have adopted a different architecture for their cloud services (e.g., master-slave replication, partitioning, distributed control and various combinations of it). As a result, the cost and performance of the services vary significantly depending on the workload. A detailed description of the architectures is provided in the paper. Furthermore, only two architectures, the one implemented on top of Amazon S3 and MS Azure using SQL Azure as the database, were able to scale and sustain our maximum workload of 9000 EBs, resulting in over 1200 Web-interactions per second (WIPS). MySQL installed on EC2 and Amazon RDS are able to sustain a maximum load of approximate 3500 EBs. MySQL Replication performed similar to MySQL standalone with EBS, so we left it off the picture. Figure 1 shows that the WIPS of Amazon’s SimpleDB grow up to about 3000 EBs and more than 200 WIPS. In fact, SimpleDB was already overloaded at about 1000 EBs and 128 WIPS in our experiments. At this point, all write requests to hot spots failed. Google AppEngine already dropped out at 500 emulated browsers with 49 WIPS. This is mainly due to Google’s transaction model not being built for such high write workloads. When implementing the benchmark, our policy was to always use the highest offered consistency guarantees, which come closest to the TPC-W requirements. Thus, in the case of AppEngine, we used the offered transaction model inside an entity group. However, it turned out, that this is a big slow-down for the whole performance. We are now in the process of re-running the experiment without transaction guarantees and curios about the new performance results.
 |
| Figure 2: Comparison of Architectures [WIPS] |
Table 1 shows the total cost per web-interaction in milli dollar for the alternative approaches and a varying load (EBs). Google AE is cheapest for low workloads (below 100 EBs) whereas Azure is cheapest for medium to large workloads (more than 100 EBs). The three MySQL variants (MySQL, MySQL/R, and RDS) have (almost) the same cost as Azure for medium workloads (EB=100 and EB=3000), but they are not able to sustain large workloads.
 |
| Table 1: Cost per WI [m$], Vary EB |
The success of Google AE for small loads has two reasons. First, Google AE is the only variant that has no fixed costs. There is only a negligible monthly fee to store the database. Second, at the time these experiments were carried out, Google gave a quota of six CPU hours per day for free. That is, applications which are below or slightly above this daily quota are particularly cheap.
Azure and the MySQL variants win for medium and large workloads because all these approaches can amortize their fixed cost for these workloads. Azure SQL server has a fixed cost per month of USD 100 for a database of up to 10 GB, independent of the number of requests that need to be processed by the database. For MySQL and MySQL/R, EC2 instances must be rented in order to keep the database online. Likewise, RDS involves an hourly fixed fee so that the cost per WIPS decreases in a load situation. It should be noted that network traffic is cheaper with Google than with both Amazon and Microsoft.
Table 2 shows the total cost per day for the alternative approaches and a varying load (EBs). (A “-” indicates that the variant was not able to sustain the load.) These results confirm the observations made previously: Google wins for small workloads; Azure wins for medium and large workloads. All the other variants are somewhere in between. The three MySQL variants come close to Azure in the range of workloads that they sustain. Azure and the three MySQL variants roughly share the same architectural principles (replication with master copy architectures). SimpleDB is an outlier in this experiment. With the current pricing scheme, SimpleDB is an exceptionally expensive service. For a large number of EBs, the high cost of SimpleDB is particularly annoying because users must pay even though SimpleDB drops many requests and is not able to sustain the workload.
Continue Reading at: High Scalability
Related posts
Cloud Computing and SaaS – Paying for Private Cloud Computing
May 11th
Perhaps the biggest challenge facing senior IT directors is how to pay for the reinvention of their enterprise IT operations as a private cloud computing service. After all, when 75 percent of the IT budget goes towards maintenance fees and existing operations, finding the money to fund new projects ranges from difficult to impossible.
In fact, Jim Ganthier, vice president of marketing for industry standard servers for Hewlett-Packard, refers to this challenge as nothing short of “IT innovation gridlock.” Unless IT organizations can find the funds to adopt the new types of systems needed to deploy a private cloud computing environment, IT will never really be managed as a true service.
Hewlett-Packard wants to help customers take on this challenge with new services that help IT organizations self-fund the transition to private cloud computing at a time when IT leaders are under pressure to innovate without increasing costs.
A research survey of more than 400 business and senior IT executives conducted by Coleman Parkes Research on behalf of HP found that one out two senior IT executives feel their company is being held back in terms of business process innovation because the majority of the IT budget was being consumed to support existing operations. The study also concludes that 95 percent of the executives survyeed said this inability to innovate had resulted in lost opportunities for the company; while 93 percent said it resulted in lost effort for the company; and 99 percent said it resulted in lost time.
Sine time and effort equals money and cost, HP thinks now is the time to roll out new cloud computing migration services while also enhancing the management functionality embedded in its Proliant G7 servers. HP is upgrading the Integrated Lights-Out (ILO) processor that the company embedds in its servers in order to add additional monitoring along with power and cooling management capabilities. The overall idea, said Ganthier, is for G7 servers to be able to pay for themselves within two months by reducing ongoing operational costs, which in turn can be used to help fund the adoption of newer technologies needed to create private cloud computing platforms.
The new services that HP wants customers to buy to help facilitate this transition include a Cloud Service Automation (CSA) offering that automates many IT processes, assessment tools for managing virtual server environments, and a variety of financial management tools designed specifically for IT organizations.
Whether IT organizations opt to lean on HP or some other vendor to make the transition to private cloud computing is up to each individual organization. But what is clear is that IT organizations are being required to modernize the way they manage IT resources in this era of the cloud. So the real question seems to be whether they are going to do it themselves; or leave the task to their eventual successor.
Related posts
Cloud Computing – Speedtests to Cloud Servers
Apr 29th
Here are some speedtests found over at: CloudHarmony
Non-US Speedtests to Cloud Servers
Global Speedtests to Cloud Servers
Test your own speeds
Related posts
Ideal Testing Tool For Cloud Computing Applications
Apr 22nd
Cloud Computing is fast emerging as a preferred technology for application development and deployment. Though the concept of Cloud Computing has been in existence for quite some time, the Technology has gained wide recognition only recently. Cloud Computing makes use of several servers placed at various locations connected to one another by high-speed network connections.
The idea governing the Cloud Computing concept is utilization of several different servers in various locations to host applications and data instead of using a single server in one location.
With Cloud Computing, consumers and businesses can access various applications without installing anything. In short, one can access any application from any computer that is equipped with an Internet connection. This is of immense benefit to businesses, in particular, as there would be no necessity of having to be at a particular computer to execute a task.
Cloud infrastructure services are offered on rent by service providers. Payment is made only for the cloud computing resources that are actually used. This is efficient as unnecessary expense is prevented.
Cloud Computing is being increasingly preferred by businesses for its many benefits, including user-friendliness, flexibility and scalability. Another increasing use of cloud-based environments is in software testing. They are being employed increasingly for software testing for their flexibility and cost-effectiveness.
This shift to Cloud environment demands Cloud-based Testing tools as conventional testing tools are ineffective in Cloud Testing. In Cloud Testing, aspects such as the network, desktop, and the impact of alterations within the Cloud, need to be focused on. As such, Testing tools employed for Cloud Testing should be so equipped that they enable software developers and testers analyze the above-mentioned aspects.
There are many Open Source software testing tools being published for Cloud environments currently. PushToTest’s TestMaker is a leading automation testing platform capable of running tests on test equipment or in a Cloud Computing environment, or both.
Full Source: Cloud Computing Applications
Related posts
A Step-by-Step Guide for Deploying eCommerce Systems in the Cloud
Apr 15th
How does one deploy cloud
computing for ecommerce systems? It’s really about the architecture, understanding your own requirements, and then understanding the cloud computing options that lay before you. Here is a quick and dirty guide for you.
Step 1: Understand the business case.
While it would seem that moving to the cloud is a technology exercise, the reality is that the core business case should be understood as to the potential benefits of cloud computing. This is the first step because there is no need to continue if we can’t make a business case. Things to consider include the value of shifting risk to the cloud computing provider, the value of on-demand scaling (which has a high value in the world of ecommerce), and the value of outsourcing versus in-sourcing.
Step 2: Understand your existing data, services, processes, and applications.
You start with what you have, and cloud computing is no exception. You need to have a data-level, service-level, and process-level understanding of your existing problem domain, also how everything is bundled into applications. I covered this in detail in my book, but the short answer is to break your existing system or systems down to a functional primitive of any architectural components, or data, services, and processes, with the intention being to assemble them as components that reside in the cloud and on-premise.
Step 3: Select a provider.
Once you understand what you need, it’s time to see where you’re going. Selecting a cloud computing provider, or, in many cases, several, is much like selecting other on-premise technologies. You line up your requirements on one side, and look at the features and functions of the providers on the other. Also, make sure to consider the soft issues such as viability in the marketplace over time, as well as security, governance, points-of-presence near your customers, and ongoing costs.
Step 4: Migrate.
In this step we migrate the right architectural assets to the cloud, including transferring and translating the data for the new environment, as well as localizing the applications, services, and processes. Migration takes a great deal of planning to pull off successfully the first time.
Step 5: Deploy.
Once your system is on the cloud computing platform, it’s time to deploy it or turn it into a production system. Typically this means some additional coding and changes to the core data, as well as standing up core security and governance systems. Moreover, you must do initial integration testing, and create any links back to on-premise systems that need to communicate with the newly deployed cloud computing systems.
Step 6: Test.
Hopefully, everything works correctly on your new cloud computing provider. Now you must verify that through testing. You need to approach this a few ways, including functional testing, or how your ecommerce system works in production, as well as performance testing, testing elasticity of scaling, security and penetration testing.
If much of this sounds like the process of building and deploying a more traditional on-premise system, you’re right. What does change, however, is that you’re not in complete control of the cloud computing provider, and that aspect of building, deploying, and managing an ecommerce system needs to be dialed into this process.
Related posts
Cloud Computing performance – Enterprise cloud put to the test
Apr 5th
The potential benefits of public clouds are obvious to most IT execs, but so are the pitfalls – outages, security concerns, compliance issues, and questions about performance, management, service-level agreements and billing. At this point, it’s fair to say that most IT execs are wary of entrusting sensitive data or important applications to the public cloud.
How we tested these cloud computing products
Archive of Network World tests
But a technology as hyped as cloud computing can’t be ignored either. IT execs are exploring the public cloud in pilot programs, they’re moving to deploy cloud principles in their own data centers, or they are eyeing an alternative that goes by a variety of names – enterprise cloud, virtual private cloud or managed private cloud.
We’re using the term enterprise cloud to mean an extension of data center resources into the cloud with the same security, audit, and management/administrative components that are best practices within the enterprise. Common use cases would be a company that wanted to add systems resources without a capital outlay during a busy time of the year or for a special, resource-intensive project or application.
In this first-of-its-kind test, we invited cloud vendors to provide us with 20 CPUs that would be used for five instances of Windows 2008 Server and five instances of Red Hat Enterprise Linux – two CPUs per instance. We also asked for a 40GB internal or SAN/iSCSI disk connection, and 1Mbps of bandwidth from our test site to the cloud provider. And we required a secure VPN connection.
Continue Reading at NetworkWorld
Related posts
Cloud Computing – Is Amazon Winning the Cloud Race?
Mar 31st
From our perspective, it looks like Amazon is winning the cloud race. Amazon and Google pioneered the notion of ‘devops‘, where agile practices are applied to merging the disciplines of development and operations. Devops teams are inherent to cloud computing. They are the only way to scale and compete.
For example, in my conversations with Amazon it’s been explained how their operations team is really only two parts:
- Infrastructure Engineering: software development for automating hardware/software and building horizontal service layers
- Datacenter Operations: rack & stack and replace broken hardware
There isn’t a team that manually configures software or hardware as in traditional operations teams or enterprise IT. This is by design. It’s the only way to effectively scale up to running thousands of servers per operator.
More importantly, by automating everything, you become fast and agile, able to build an ecosystem of cloud services more rapidly than your competitors.
With the possible exception of Rackspace Cloud, I’m not sure that anyone else is in Amazon or Google’s league. Amazon in particular, has a track record that is incredibly impressive.
From a quick culling of all of the Amazon Web Services press releases since the launch of it’s initial service (SQS in 2004), after removing non-feature press releases and minor releases of little value, we came up with the following graph:
The trend is clear. Since Amazon’s start, they have accelerated rapidly, almost doubling their feature releases every year. 2009 was spectacular with 43 feature releases of note. Since the beginning of 2010, Amazon already has 8 releases of note.
In contrast, traditional hosting companies moving into cloud computing are hobbled by running two teams: development and operations. Expect the gap to widen as more hosting companies continue to misunderstand that this race isn’t about technology; it’s about people, software, and discipline.
So what’s the takeaway? Simply put, in order to be a major cloud player you need to change how you do IT and build clouds. Either hire someone who can bring the devops practice into your shop or engage an cloud computing engineering services firm like Cloudscaling to help you build fast, nimble teams that focus on automation and rapid release cycles. Full Source CloudScaling
Related posts
Next-Generation Data Centers – Cloud Computing
Mar 15th
Flexibility in design and a more modular approach could extend the life–and usefulness–of data centers.
Cloud computing may be getting the headlines, but that hasn’t diminished interest in building new data centers.
The new data centers are being designed differently, though. They’re more modular, more flexible and much more accepting of new technology as it becomes available. The idea of building a data center every decade has met with the economic reality inside most companies. It’s simply too expensive.
So what exactly is different? Forbes sat down with Steve Sams, vice president of site and facilities services at IBM ( IBM – news – people ), to find out.
Forbes: What’s happening inside data centers these days?
Steve Sams: More than 75% of our customers are experiencing major challenges. Their IT growth continues to explode and they’re trying to run all of this in data centers that are up to 20 years old. The average age of a data center, according to a recent IDC study, is nine years old. Gartner suggests any data center more than seven years old is obsolete.
Why does it become obsolete so quickly?
Technology price/performance improvements are delivered by getting things into smaller and smaller footprints. Technology becomes more dense. The amount of power and cooling required if the rack is fully loaded has been climbing steadily. ASHRAE (American Society of Heating, Refrigerating and Air Conditioning Engineers) estimates that over the last decade technology density has increased 20 times. If you build a data center to support racks that are running at 1,000 watts per rack and you’re installing blade technology that’s running at 20,000 watts per rack, you have a mismatch. That’s occurring inside data centers all around the world.
Isn’t that one of the reasons companies are migrating to clouds?
We’re certainly seeing a lot of business for data servicing companies, where we’re building data centers for them, as well as for companies that are in-sourcing their cloud environment–or continuing to grow in a more virtualized way. In countries with more of a capital crunch the cloud is very popular. In countries like China and India where the customers are more focused on owning the assets themselves, they are interested in owning the data centers. There’s been a huge mismatch between data centers and technology. Technology historically got refreshed every three to five years while data centers were designed for 20 years. Continue Reading at Forbes
