Posts tagged database
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
Sneaking around IT to get to the cloud – Cloud Computing Blog
Apr 23rd
A Ponemon Institute survey recently piqued my interest. In the 2010 Access Governance Trends Survey, 87 percent of respondents said too many employees were able to access information that should have been out of reach. And guess what? Cloud computing was a factor — 73 percent of respondents said that cloud-based applications were enabling business users to skirt organizational controls.
The core issue is IT’s loss of control over its assets, including data. Let’s face it — departments are sick of waiting for development and deployment of core business applications or infrastructure services, and they’re going directly to a cloud computing provider to get what they need. Think of it as a kind of technological infidelity.
[ Get the no-nonsense explanations and advice you need to take real advantage of cloud computing in the InfoWorld's 21-page Cloud Computing Deep Dive PDF special report, featuring an exclusive excerpt from David Linthicum's new book on cloud architecture. | Stay up on the cloud with InfoWorld's Cloud Computing Report newsletter. ]
Going around IT and straight to the cloud has become common practice in the last few years. Salesforce.com built its business selling directly to the sales staff rather than to IT; eventually, IT was forced to accept SaaS (software as a service) after the fact. I’ve watched those battles firsthand.
Today, things are even worse. Now you can get storage as a service, database as a service, and even complete application servers and app dev platforms that are delivered on-demand. With such endless resources available, corporate fiefdoms are creating so-called rogue clouds — their own array of cloud computing services, including data repositories, that they alone control. IT may not have a clue about what’s going on.
The trouble with the rogue approach is that there’s no way to ensure that data is handled in accordance with corporate policies. Worse, that data may come with compliance issues, including personal medical or financial information where the law dictates how the data is handled and where it can reside.
Full Source: InfoWorld
Related posts
Verizon’s On-Demand Cloud Computing Solution Adds Server Cloning, Increases Flexibility
Apr 15th
(WEB HOST INDUSTRY REVIEW) — Based on customer demand, US telecom Verizon (www.verizon.com) has enhanced its global, on-demand Computing as a Service offering (known as CaaS), providing business customers with better control and flexibility over their computing environments.
According to its Thursday announcement, CaaS clients now have access to server cloning, application and operating system expansion, and expanded networking flexibility.
Server cloning gives IT administrators the option to customize the configuration of a CaaS virtual server and then create a golden, or reference, server image, eliminating the need to manually create the same server image multiple times and lets them rapidly deploy server clones supporting the same corporate application.
With added support for SUSE Linux, the Verizon CaaS platform can be used with commonly deployed enterprise resource planning packages. In addition, Microsoft SQL Server 2008 has been added as a “click-to-provision” database server option.
And enterprises are given expanded networking flexibility with virtual router and shared virtual private options including Verizon Private IP for connecting back-end systems to Verizon CaaS via an online portal. In addition, customers can purchase metered, burstable bandwidth up to 1Gbps to meet immediate requirements for temporary computing capacity.
In addition to adding new features, Verizon has successfully completed the first annual SAS 70 Type II examination of controls for its CaaS data centers, demonstrating that Verizon Business has controls and processes in place to manage and monitor its CaaS platform, as well as customers’ critical applications and infrastructure. The SAS70 Type II is part of the Sarbanes-Oxley Act that governs data center operations and internal controls, and while it was originally intended for the financial services industry, the SAS70 has become a widely recognized auditing standard for evaluating outsourced data center solutions. This evaluation helps to ease concerns customers may have about migrating the delivery of critical IT services to the cloud.
“Verizon CaaS was widely recognized as one of the industry’s most comprehensive cloud-based computing solution for enterprises when it was launched last June,” Verizon Business IT solutions executive director Joseph S. Crawford said in a statement. “With the addition of new features and the completion of the SAS 70 Type II data center audit, we are continuing the investment in our flagship cloud offering as enterprise clients increasingly look to cloud computing to fundamentally change the way they consume IT resources.” Full Source: TheWhir
Related Blogs
- Verizon’s On-Demand Cloud Computing Solution Adds Server Cloning …
- Grid Computing in Distributed Gis | Technology Base
- The Reinvention of the Cloud Computing Reseller
- Frustrations with cloud computing mount
Related posts
The Last Mile’ Of Virtualization And Cloud Computing In Retail
Apr 5th
There is no question, virtualization is here to stay and it is one of the most significant enabling technologies for efficiently delivering cloud computing. “So what?” you might ask. Whether you are the consumer of, or a supplier of computing technology within retail, cloud computing is the most significant technology game changer to hit the industry since scanning, and you should care because it is going to change the way IT gets done in this industry. You need to be ready.
Quite a prediction. What does it mean? Some may look at this whole area of cloud computing and say, “What is old is new again,” because we are simply going back to the service bureaus of days gone by. While on some level that is true, the benefits are far greater this time around. Whether you call it cloud computing or SaaS (software as a service) or utility-based computing or any number of near-synonyms, the concept has been around for quite some time but has never quite been perfected, and the reasons for that vary.
Numerous organizations offer an ASP model re-branded as SaaS, but were never able to take it all the way to a utility model. Others provide cloud computing but only for a portion of what the retailer needs. Others will argue that only certain components of a retailer’s suite should be outsourced as SaaS. Many discussion groups in our new social mediums have started up on this topic and they are full of folks talking about RIA (rich Internet applications), SOA (service oriented architecture), smart clients, virtualization at POS —you name the buzz word and it’s in there — all of which circle around the topic of the future of cloud computing but miss the mark because they focus on the how, not the why. What Gartner is saying in their quote above — and this is the crux of the paradigm shift that cloud computing creates — is that organizations are going to get out of the IT business. Why? Because there is no longer any strategic value in owning and operating your own IT. So given that, it would follow that if a retailer subscribes to this notion, they don’t care how the service gets delivered, just that their business needs are met. It’s like when you subscribe to an online CRM system. You don’t care what brand of routers are As seen in the 04/05/10 edition of the Retail Solutions Online (www.retailsolutionsonline.com) newsletter. used or which backend database everything resides on or what hardware it runs on and how you keep that current. You just care that your business needs are met. Full Source
Related posts
Announcing Combined AWS Data Transfer Pricing – Amazon Cloud Computing
Mar 26th
Starting April 1, 2010, your Data Transfer Out pricing tier for a given Region will be based on your total Data Transfer Out usage within that Region for Amazon Simple Storage Service (Amazon S3), Amazon Elastic Compute Cloud (Amazon EC2), Amazon SimpleDB, Amazon Relational Database Service (Amazon RDS), Amazon Virtual Private Cloud (Amazon VPC), and Amazon Simple Queue Service (Amazon SQS). Until now, usage tiers have been calculated individually for each service, based on data transfer related to that service. Because AWS is now aggregating your total Data Transfer Out usage across multiple services, you can reach higher usage tiers and lower pricing more quickly. In addition, you’ll benefit from a complimentary tier which provides your first GB of outbound transfer in each Region each month at no charge.
The tiered pricing for Data Transfer Out is as follows for each Region:
- First 1 GB of data transferred out per month is free
- Remainder of first 10 TB per Month: $0.15 per GB
- Next 40 TB per Month: $0.11 per GB
- Next 100 TB per Month: $0.09 per GB
- Over 150 TB per Month: $0.08 per GB
As you may know, all inbound data transfer is free of charge until June 30, 2010. All data transfer usage (both inbound and outbound) for participating Amazon Web Services now appears in aggregate in its own section of your AWS account activity page and monthly bill. As a bonus, you’ll notice that your first GB of outbound data transfer in each Region is now included free of charge.
Source Amazon Website
Related posts
Cloud computing: the biggest threats
Mar 2nd
The next generation of cloud-based storage and services allows us greater freedom to compute on the go. It enables us to be free from the storage constraints of compact devices and gives us the ability to access services and personal information on these devices regardless of our location. Cloud computing also poses serious threats to online privacy.
A new report published by The Cloud Security Alliance (CSA) and HP on March 1 details some of the more serious threats about what could occur when using cloud-based services, now and in the future.
Perhaps the most worrisome threat for individuals is the loss and interception of personal information such as credit card details, banking records, medical records, home and work address and any other information that could leave them susceptible to identity theft.
A lot of research has gone into creating cloud-based services, but perhaps not enough time and money has been spent protecting individuals’ privacy and safeguarding against the interception of information via third-party sites. More resources need to be harnessed to ensure these services will be safe for people to use when future generations of hackers try to prey on cloud-based databases.
There are few laws in place that govern cloud-based security practices. Cloud computing privacy policies are often very vague about what happens in the event of information loss or theft.
This level of uncertainty is largely replicated in the early adoption of cloud computing. Security is cited as the number one barrier to adoption. New users find it difficult to weigh up the pros and cons of cloud computing; there is a wealth of opportunities floating in the cloud but customers are very concerned about the associated risks.
By highlighting the top cloud computing security issues, CSA and HP hope to make cloud computing safer for consumers and businesses alike.
“Cloud services are clearly the next generation of information technology that enterprises must master. We have a shared responsibility to understand the security threats that accompany the cloud and apply the necessary best practices to mitigate them,” said Jim Reavis, founder of the Cloud Security Alliance.
The top security threats of cloud computing:
1. Abuse and Nefarious Use of Cloud Computing
2. Insecure Interfaces and APIs
3. Malicious Insiders
4. Shared Technology Issues
5. Data Loss or Leakage
6. Account or Service Hijacking
7. Unknown Risk Profile
Related posts
Why the public sector is opening the doors to open source technology?
Jul 16th
With the public sector spending plans of all political parties coming under close scrutiny, IT is being labelled in equal measure as the cause and saviour of the current spending crunch that the public sector is facing.
However, it is not a simple either/or argument. It is not a case of should the country have an eGovernment strategy, but rather what that strategy should be.
The debate reached a new height this week with the Conservative Party’s announcement regarding its suggested alternative to the spiralling costs of the NHS database project. Moving on from general statements of intent, this was the one of the first acknowledgements that government technology purchasing is under the spotlight and seen to be an important part of a party’s manifesto.
The idea of moving NHS data into the ‘cloud’ through providers such as Google is a bold choice; however it mimics strategies that have been adopted across the private sector. There is no doubt that the landscape of public sector technology purchasing is changing and the adoption of open source/ open standards technology provides a more transparent, democratic approach to technology purchasing.
