Testing the Cloud: Assuring Availability

By Joe Barry

August 16, 2011

Cloud computing is changing how IT services are delivered and consumed today. The ability for enterprises large and small to centralize and outsource increasingly complex IT infrastructure, while at the same time consuming cloud-based IT services on an on-demand basis, promises to transform the economics of doing business.

But, note that I state “promises” as even though there are many success stories amongst early adopters, the real test will come when cloud computing becomes the de-facto model for IT service delivery and consumption. By all accounts, mainstream adoption of cloud services is close at hand.

In essence, cloud computing is entering a new phase in its development, where assuring the availability and quality of cloud services will become a major challenge. Preparing for this now will ensure that cloud computing continues to deliver on its “promise”.

From excess to scarce resources

Cloud computing was initially driven by excess computing capacity. Large web companies, such as Amazon and Google, that had to build large data center capacity for their own business, saw an opportunity to provide their excess capacity as a service to others. This has been so successful that these cloud services, such as Amazon Web Services, have become a business in themselves.

Yet, as these services become more popular, demand will tend to outstrip supply, especially as some of the enablers of cloud service adoption, such as higher speed access connections, continue to grow in capacity. Simply adding more servers and higher speed networks is effective, but costly and can undermine one of the main reasons for using cloud services, namely cost reduction.  Cloud service providers will thus face the dilemma of managing demand for scarcer computing resources while at the same time maintaining a low, or at least competitive, cost level.

In other words, how can cloud service providers meet mainstream demand cost-effectively?

Efficiently Assuring Service Availability

Cloud services come in many shapes and sizes, from private clouds to public clouds with software-, platform- and infrastructure-as-a-service. Nevertheless, all these flavors of cloud service have a common need to assure service availability and do so as efficiently and cost effectively as possible.

Many cloud services already provide service availability monitoring tools, but these are often limited to monitoring of server or service up-time. Server or service up-time is but one of the aspects of service availability that need to be addressed as cloud services are dependent on much more than just the physical or virtual server on which they reside. Increasingly, the data communication infrastructure supporting the cloud service from the provider to the consumer also needs to be assured even though this might be outside the direct control of the service provider.  

To ensure mainstream adoption of cloud services, consumers must be confident that the services that are required or the data that is hosted by cloud services is available quickly when and where they need it. Otherwise, why not continue with current approaches? Mainstream consumers are noted for being more conservative and pragmatic in their choice of solutions, so addressing this concern must be a top priority for continued expansion of cloud service adoption.

Therefore, building the infrastructure to test and monitor cloud services is essential.

Testing and monitoring cloud services

From a testing and monitoring perspective, there are a number of layers one can address:

•    The Wide Area Network (WAN) providing data communication services between the enterprise customer and the cloud service – fundamental to service assurance and testing of end-to-end service availability

•    The data center infrastructure comprising servers and data communication between servers (LAN), where service availability and uptime of this equipment is key as well as efficient use of resources to ensure service efficiency

•    The monitoring infrastructure in the data center that is the basis for service assurance which itself needs to efficient

•    The individual servers and monitoring appliances that are based on servers that must also follow efficiency and availability principles to assure overall service efficiency and service availability
 
Testing end-to-end

The first test that can be performed is testing end-to-end availability. At a basic level, this involves testing connectivity, but can also involve some specific testing relevant for cloud services, such as latency measurement. Several commercial systems exist for testing latency in a WAN environment. These are most often used by financial institutions to determine the time it takes to execute financial transactions with remote stock exchanges, but can also be used by cloud service providers to test the latency of the connection to enterprise customers.

This solution requires the installation at the enterprise of a network appliance for monitoring latency, which could also be used to test connectivity. Such an appliance could also be used for troubleshooting and SLA monitoring.

Typically the cloud service provider does not own the WAN data communication infrastructure. However, using network monitoring and analysis appliances at both the data center and the enterprise, it is possible to measure the performance of the WAN in providing the data communication service required. The choice of WAN data communication provider should also be driven by the ability of this provider to provide performance data in support of agreed SLAs. In other words, this provider should have the monitoring and analysis infrastructure in place to assure services.

From reaction to service assurance

Network monitoring and analysis of the data center infrastructure is also crucial as cloud service providers need to rely less on troubleshooting and more on service assurance strategies. In typical IT network deployments, a reactive strategy is preferred whereby issues are dealt with in a troubleshooting manner as they arise. For enterprise LAN environments, this can be acceptable in many cases, as some downtime can be tolerated. However, for cloud service providers, downtime is a disaster! If customers are not confident in the cloud service provider’s ability to assure service availability, they will be quick to find alternatives or even revert to a local installation.

A service assurance strategy involves constant monitoring of the performance of the network and services so that issues can be identified before they arise. Network and application performance monitoring tools are available from a number of vendors for precisely this purpose.

The power of virtualization

One of the technology innovations of particular use to cloud service providers is virtualization. The ability to consolidate multiple cloud services onto as few physical servers as possible provides tremendous efficiency benefits by lower cost, space and power consumption. In addition, the ability to move virtual machines supporting cloud services from one physical server to another allows efficient use of resources in matching time-of-day demand, as well as allowing fast reaction to detected performance issues.

One of the consequences of this consolidation is the need for higher speed interfaces as more data needs to be delivered to each server. This, in turn requires that the data communication infrastructure is dimensioned to provide this data, which in turn demands that the network monitoring infrastructure can keep up with the data rates without losing data. This is far from a given, so cloud service providers need to pay particular attention to the throughput performance of network monitoring and analysis appliances to ensure that they can keep up also in the future.

Within the virtualized servers themselves, there are also emerging solutions to assist in monitoring performance. Just as network and application performance monitoring appliances are available to monitor the physical infrastructure, there are now available virtualized versions of these applications for monitoring virtual applications and communication between virtual machines.

There are also virtual test applications that allow a number of virtual ports to be defined that can be used for load-testing in a cloud environment. This is extremely useful for testing whether a large number of users can access a service without having to deploy a large test network. An ideal tool for cloud service providers.

Bringing virtualization to network monitoring and analysis

While virtualization has been used to improve service efficiency, the network monitoring and analysis infrastructure is still dominated by single server implementations. In many cases, this is because the network monitoring and analysis appliance requires all the processing power it can get. However, there are opportunities to consolidate appliances, especially as servers and server CPUs increase performance on a yearly basis.

Solutions are now available to allow multiple network monitoring and analysis applications to be hosted on the same physical server. If all the applications are based on the same operating system, intelligent network adapters have the ability to ensure that data is shared between these applications, which often need to analyze the same data at the same time, but for different purposes.

However, for situations where the applications are based on different operating systems, virtualization can be used to consolidate them onto a single physical server. Demonstrations have shown that up to 32 applications can thus be consolidated using virtualization.

By pursuing opportunities for consolidation of network monitoring and analysis appliances, cloud service providers can further improve service efficiency.

Preparing for mainstream adoption

Mainstream adoption of cloud services is just around the corner and to take full advantage of this demand, cloud service providers can use the existing tools and concepts described above to assure service availability in a cost effective and efficient manner.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

Fluid HPC: How Extreme-Scale Computing Should Respond to Meltdown and Spectre

February 15, 2018

The Meltdown and Spectre vulnerabilities are proving difficult to fix, and initial experiments suggest security patches will cause significant performance penalties to HPC applications. Even as these patches are rolled o Read more…

By Pete Beckman

Intel Touts Silicon Spin Qubits for Quantum Computing

February 14, 2018

Debate around what makes a good qubit and how best to manufacture them is a sprawling topic. There are many insistent voices favoring one or another approach. Referencing a paper published today in Nature, Intel has offe Read more…

By John Russell

Brookhaven Ramps Up Computing for National Security Effort

February 14, 2018

Last week, Dan Coats, the director of Director of National Intelligence for the U.S., warned the Senate Intelligence Committee that Russia was likely to meddle in the 2018 mid-term U.S. elections, much as it stands accused of doing in the 2016 Presidential election. Read more…

By John Russell

HPE Extreme Performance Solutions

Safeguard Your HPC Environment with the World’s Most Secure Industry Standard Servers

Today’s organizations operate in an environment with ever-evolving threats, and in order to protect themselves they must continuously bolster their security strategy. Hewlett Packard Enterprise (HPE) and Intel® are addressing modern security challenges with the world’s most secure industry standard servers powered by the latest generation of Intel® Xeon® Scalable processors. Read more…

AI Cloud Competition Heats Up: Google’s TPUs, Amazon Building AI Chip

February 12, 2018

Competition in the white hot AI (and public cloud) market pits Google against Amazon this week, with Google offering AI hardware on its cloud platform intended to make it easier, faster and cheaper to train and run machi Read more…

By Doug Black

Fluid HPC: How Extreme-Scale Computing Should Respond to Meltdown and Spectre

February 15, 2018

The Meltdown and Spectre vulnerabilities are proving difficult to fix, and initial experiments suggest security patches will cause significant performance penal Read more…

By Pete Beckman

Brookhaven Ramps Up Computing for National Security Effort

February 14, 2018

Last week, Dan Coats, the director of Director of National Intelligence for the U.S., warned the Senate Intelligence Committee that Russia was likely to meddle in the 2018 mid-term U.S. elections, much as it stands accused of doing in the 2016 Presidential election. Read more…

By John Russell

AI Cloud Competition Heats Up: Google’s TPUs, Amazon Building AI Chip

February 12, 2018

Competition in the white hot AI (and public cloud) market pits Google against Amazon this week, with Google offering AI hardware on its cloud platform intended Read more…

By Doug Black

Russian Nuclear Engineers Caught Cryptomining on Lab Supercomputer

February 12, 2018

Nuclear scientists working at the All-Russian Research Institute of Experimental Physics (RFNC-VNIIEF) have been arrested for using lab supercomputing resources to mine crypto-currency, according to a report in Russia’s Interfax News Agency. Read more…

By Tiffany Trader

The Food Industry’s Next Journey — from Mars to Exascale

February 12, 2018

Global food producer and one of the world's leading chocolate companies Mars Inc. has a unique perspective on the impact that exascale computing will have on the food industry. Read more…

By Scott Gibson, Oak Ridge National Laboratory

Singularity HPC Container Start-Up – Sylabs – Emerges from Stealth

February 8, 2018

The driving force behind Singularity, the popular HPC container technology, is bringing the open source platform to the enterprise with the launch of a new vent Read more…

By George Leopold

Dell EMC Debuts PowerEdge Servers with AMD EPYC Chips

February 6, 2018

AMD notched another EPYC processor win today with Dell EMC’s introduction of three PowerEdge servers (R6415, R7415, and R7425) based on the EPYC 7000-series p Read more…

By John Russell

‘Next Generation’ Universe Simulation Is Most Advanced Yet

February 5, 2018

The research group that gave us the most detailed time-lapse simulation of the universe’s evolution in 2014, spanning 13.8 billion years of cosmic evolution, is back in the spotlight with an even more advanced cosmological model that is providing new insights into how black holes influence the distribution of dark matter, how heavy elements are produced and distributed, and where magnetic fields originate. Read more…

By Tiffany Trader

Inventor Claims to Have Solved Floating Point Error Problem

January 17, 2018

"The decades-old floating point error problem has been solved," proclaims a press release from inventor Alan Jorgensen. The computer scientist has filed for and Read more…

By Tiffany Trader

Japan Unveils Quantum Neural Network

November 22, 2017

The U.S. and China are leading the race toward productive quantum computing, but it's early enough that ultimate leadership is still something of an open questi Read more…

By Tiffany Trader

AMD Showcases Growing Portfolio of EPYC and Radeon-based Systems at SC17

November 13, 2017

AMD’s charge back into HPC and the datacenter is on full display at SC17. Having launched the EPYC processor line in June along with its MI25 GPU the focus he Read more…

By John Russell

Researchers Measure Impact of ‘Meltdown’ and ‘Spectre’ Patches on HPC Workloads

January 17, 2018

Computer scientists from the Center for Computational Research, State University of New York (SUNY), University at Buffalo have examined the effect of Meltdown Read more…

By Tiffany Trader

Nvidia Responds to Google TPU Benchmarking

April 10, 2017

Nvidia highlights strengths of its newest GPU silicon in response to Google's report on the performance and energy advantages of its custom tensor processor. Read more…

By Tiffany Trader

IBM Begins Power9 Rollout with Backing from DOE, Google

December 6, 2017

After over a year of buildup, IBM is unveiling its first Power9 system based on the same architecture as the Department of Energy CORAL supercomputers, Summit a Read more…

By Tiffany Trader

Fast Forward: Five HPC Predictions for 2018

December 21, 2017

What’s on your list of high (and low) lights for 2017? Volta 100’s arrival on the heels of the P100? Appearance, albeit late in the year, of IBM’s Power9? Read more…

By John Russell

Russian Nuclear Engineers Caught Cryptomining on Lab Supercomputer

February 12, 2018

Nuclear scientists working at the All-Russian Research Institute of Experimental Physics (RFNC-VNIIEF) have been arrested for using lab supercomputing resources to mine crypto-currency, according to a report in Russia’s Interfax News Agency. Read more…

By Tiffany Trader

Leading Solution Providers

Chip Flaws ‘Meltdown’ and ‘Spectre’ Loom Large

January 4, 2018

The HPC and wider tech community have been abuzz this week over the discovery of critical design flaws that impact virtually all contemporary microprocessors. T Read more…

By Tiffany Trader

Perspective: What Really Happened at SC17?

November 22, 2017

SC is over. Now comes the myriad of follow-ups. Inboxes are filled with templated emails from vendors and other exhibitors hoping to win a place in the post-SC thinking of booth visitors. Attendees of tutorials, workshops and other technical sessions will be inundated with requests for feedback. Read more…

By Andrew Jones

How Meltdown and Spectre Patches Will Affect HPC Workloads

January 10, 2018

There have been claims that the fixes for the Meltdown and Spectre security vulnerabilities, named the KPTI (aka KAISER) patches, are going to affect applicatio Read more…

By Rosemary Francis

GlobalFoundries, Ayar Labs Team Up to Commercialize Optical I/O

December 4, 2017

GlobalFoundries (GF) and Ayar Labs, a startup focused on using light, instead of electricity, to transfer data between chips, today announced they've entered in Read more…

By Tiffany Trader

Tensors Come of Age: Why the AI Revolution Will Help HPC

November 13, 2017

Thirty years ago, parallel computing was coming of age. A bitter battle began between stalwart vector computing supporters and advocates of various approaches to parallel computing. IBM skeptic Alan Karp, reacting to announcements of nCUBE’s 1024-microprocessor system and Thinking Machines’ 65,536-element array, made a public $100 wager that no one could get a parallel speedup of over 200 on real HPC workloads. Read more…

By John Gustafson & Lenore Mullin

Flipping the Flops and Reading the Top500 Tea Leaves

November 13, 2017

The 50th edition of the Top500 list, the biannual publication of the world’s fastest supercomputers based on public Linpack benchmarking results, was released Read more…

By Tiffany Trader

V100 Good but not Great on Select Deep Learning Aps, Says Xcelerit

November 27, 2017

Wringing optimum performance from hardware to accelerate deep learning applications is a challenge that often depends on the specific application in use. A benc Read more…

By John Russell

2017 Gordon Bell Prize Finalists Named

October 23, 2017

The three finalists for this year’s Gordon Bell Prize in High Performance Computing have been announced. They include two papers on projects run on China’s Read more…

By John Russell

  • arrow
  • Click Here for More Headlines
  • arrow
Share This