PT-Logo
Forgot your password?
BenchmarkXPRT Blog banner

Category: Performance benchmarking

CloudXPRT is on the way

A few months ago, we wrote about the possibility of creating a datacenter XPRT. In the intervening time, we’ve discussed the idea with folks both in and outside of the XPRT Community. We’ve heard from vendors of datacenter products, hosting/cloud providers, and IT professionals that use those products and services.

The common thread that emerged was the need for a cloud benchmark that can accurately measure the performance of modern, cloud-first applications deployed on modern infrastructure as a service (IaaS) platforms, whether those platforms are on-premises, hosted elsewhere, or some combination of the two (hybrid clouds). Regardless of where clouds reside, applications are increasingly using them in latency-critical, highly available, and high-compute scenarios.

Existing datacenter benchmarks do not give a clear indication of how applications will perform on a given IaaS infrastructure, so the benchmark should use cloud-native components on the actual stacks used for on-prem and public cloud management.

We are planning to call the benchmark CloudXPRT. Our goal is for CloudXPRT to address the needs described above while also including the elements that have made the other XPRTs successful. We plan for CloudXPRT to

  • Be relevant to on-prem (datacenter), private, and public cloud deployments
  • Run on top of cloud platform software such as Kubernetes
  • Include multiple workloads that address common scenarios like web applications, AI, and media analytics
  • Support multi-tier workloads
  • Report relevant metrics including both throughput and critical latency for responsiveness-driven applications and maximum throughput for applications dependent on batch processing

CloudXPRT’s workloads will use cloud-native components on an actual stack to provide end-to-end performance metrics that allow users to choose the best IaaS configuration for their business.

We’ve been building and testing preliminary versions of CloudXPRT for the last few months. Based on the progress so far, we are shooting to have a Community Preview of CloudXPRT ready in mid- to late-March with a version for general availability ready about two months later.

Over the coming weeks, we’ll be working on getting out more information about CloudXPRT and continuing to talk with interested parties about how they can help. We’d love to hear what workflows would be of most interest to you and what you would most like to see in a datacenter/cloud benchmark. Please feel free to contact us!

Bill

AIXPRT’s unique development path

With four separate machine learning toolkits on their own development schedules, three workloads, and a wide range of possible configurations and use cases, AIXPRT has more moving parts than any of the XPRT benchmark tools to date. Because there are so many different components, and because we want AIXPRT to provide consistently relevant evaluation data in the rapidly evolving AI and machine learning spaces, we anticipate a cadence of AIXPRT updates in the future that will be more frequent than the schedules we’ve used for other XPRTs in the past. With that expectation in mind, we want to let AIXPRT testers know that when we release an AIXPRT update, they can expect minimized disruption, consideration for their testing needs, and clear communication.

Minimized disruption

Each AIXPRT toolkit (Intel OpenVINO, TensorFlow, NVIDIA TensorRT, and Apache MXNet) is on its own development schedule, and we won’t always have a lot of advance notice when new versions are on the way. Hypothetically, a new version of OpenVINO could release one month, and a new version of TensorRT just two months later. Thankfully, the modular nature of AIXPRT’s installation packages ensures that we won’t need to revise the entire AIXPRT suite every time a toolkit update goes live. Instead, we’ll update each package individually when necessary. This means that if you only test with a single AIXPRT package, updates to the other packages won’t affect your testing. For us to maintain AIXPRT’s relevance, there’s unfortunately no way to avoid all disruption, but we’ll work to keep it to a minimum.

Consideration for testers

As we move forward, when software compatibility issues force us to update an AIXPRT package, we may discover that the update has a significant effect on results. If we find that results from the new package are no longer comparable to those from previous tests, we’ll share the differences that we’re seeing in our lab. As always, we will use documentation and versioning to make sure that testers know what to expect and  that there’s no confusion about which package to use.

Clear communication

When we update any package, we’ll make sure to communicate any updates in the new build as clearly as possible. We’ll document all changes thoroughly in the package readmes, and we’ll talk through significant updates here in the blog. We’re also available to answer questions about AIXPRT and any other XPRT-related topic, so feel free to ask!

Justin

Planning for the next CrXPRT

We’re currently planning the next version of CrXPRT, our benchmark that evaluates the performance and battery life of Chromebooks. If you’re unfamiliar with CrXPRT, you can find out more about how it works both here in the blog and at CrXPRT.com. If you’ve used CrXPRT, we’d love to hear any suggestions you may have. What do you like or dislike about CrXPRT? What features do you hope to see in a new version?

When we begin work on a new version of any benchmark, one of our first steps is to determine whether the workloads will provide value during the years ahead. As technology and user behavior evolve, we update test content to be more relevant. One example is when we replace photos with ones that use more contemporary file resolutions and sizes.

Sometimes the changing tech landscape prompts us to remove entire workloads and add new ones. The Photo Collage workload in CrXPRT uses Portable Native Client (PNaCl) technology, for which the Chrome team will soon end support. CrXPRT 2015 has a workaround for this issue, but the best course of action for the next version of CrXPRT will be to remove this workload altogether.

The battery life test will also change. Earlier this year, we started to see unusual battery life estimates and high variance when running tests at CrXPRT’s default battery life test length of 3.5 hours, so we’ve been recommending that users perform full rundowns instead. In the next CrXPRT, the battery life test will require full rundowns.

We’ll also be revamping the CrXPRT UI to improve the look of the benchmark and make it easier to use, as we’ve done with the other recent XPRT releases.

We really do want to hear your ideas, and any feedback you send has a chance to shape the future of the benchmark. Let us know what you think!

Justin

The XPRT Spotlight Black Friday Showcase helps you shop with confidence

Black Friday and Cyber Monday are almost here, and you may be feeling overwhelmed by the sea of tech gifts to choose from. The XPRTs are here to help. We’ve gathered the product specs and performance facts for some of the hottest tech devices in one convenient place—the XPRT Spotlight Black Friday Showcase. The Showcase is a free shopping tool that provides side-by-side comparisons of some of the season’s most popular smartphones, laptops, Chromebooks, tablets, and PCs. It helps you make informed buying decisions so you can shop with confidence this holiday season.

Want to know how the Google Pixel 4 stacks up against the Apple iPhone 11 or Samsung Galaxy Note10 in web browsing performance or screen size? Simply select any two devices in the Showcase and click Compare. You can also search by device type if you’re interested in a specific form factor such as consoles or tablets.

The Showcase doesn’t go away after Black Friday. We’ll rename it the XPRT Holiday Showcase and continue to add devices such as the Microsoft Surface Pro X throughout the shopping season. Be sure to check back in and see how your tech gifts measure up.

If this is the first you’ve heard about the XPRT Tech Spotlight, here’s a little background. Our hands-on testing process equips consumers with accurate information about how devices function in the real world. We test devices using our industry-standard BenchmarkXPRT tools: WebXPRT, MobileXPRT, TouchXPRT, CrXPRT, BatteryXPRT, and HDXPRT. In addition to benchmark results, we include photographs, specs, and prices for all products. New devices come online weekly, and you can browse the full list of almost 200 that we’ve featured to date on the Spotlight page.

If you represent a device vendor and want us to feature your product in the XPRT Tech Spotlight, please visit the website for more details.

Justin

Understanding AIXPRT’s default number of requests

A few weeks ago, we discussed how AIXPRT testers can adjust the key variables of batch size, levels of precision, and number of concurrent instances by editing the JSON test configuration file in the AIXPRT/Config directory. In addition to those key variables, there is another variable in the config file called “total_requests” that has a different default setting depending on the AIXPRT test package you choose. This setting can significantly affect a test run, so it’s important for testers to know how it works.

The total_requests variable specifies how many inference requests AIXPRT will send to a network (e.g., ResNet-50) during one test iteration at a given batch size (e.g., Batch 1, 2, 4, etc.). This simulates the inference demand that the end users place on the system. Because we designed AIXPRT to run on different types of hardware, it makes sense to set the default number of requests for each test package to suit the most likely hardware environment for that package.

For example, testing with OpenVINO on Windows aligns more closely with a consumer (i.e., desktop or laptop) scenario than testing with OpenVINO on Ubuntu, which is more typical of server/datacenter testing. Desktop testers require a much lower inference demand than server testers, so the default total_requests settings for the two packages reflect that. The default for the OpenVINO/Windows package is 500, while the default for the OpenVINO/Ubuntu package is 5,000.

Also, setting the number of requests so low that a system finishes each workload in less than 1 second can produce high run-to-run variation, so our default settings represent a lower boundary that will work well for common test scenarios.

Below, we provide the current default total_requests setting for each AIXPRT test package:

  • MXNet: 1,000
  • OpenVINO Ubuntu: 5,000
  • OpenVINO Windows: 500
  • TensorFlow Ubuntu: 100
  • TensorFlow Windows: 10
  • TensorRT Ubuntu: 5,000
  • TensorRT Windows: 500


Testers can adjust these variables in the config file according to their own needs. Finding the optimal combination of machine learning variables for each scenario is often a matter of trial and error, and the default settings represent what we think is a reasonable starting point for each test package.

To adjust the total_requests setting, start by locating and opening the JSON test configuration file in the AIXPRT/Config directory. Below, we show a section of the default config file (CPU_INT8.json) for the OpenVINO-Windows test package (AIXPRT_1.0_OpenVINO_Windows.zip). For each batch size, the total_requests setting appears at the bottom of the list of configurable variables. In this case, the default setting Is 500. Change the total_requests numerical value for each batch size in the config file, save your changes, and close the file.

Total requests snip

Note that if you are running multiple concurrent instances, OpenVINO and TensorRT automatically distribute the number of requests among the instances. MXNet and TensorFlow users must manually allocate the instances in the config file. You can find an example of how to structure manual allocation here. We hope to make this process automatic for all toolkits in a future update.

We hope this information helps you understand the total_requests setting, and why the default values differ from one test package to another. If you have any questions or comments about this or other aspects of AIXPRT, please let us know.

Justin

AIXPRT is here!

We’re happy to announce that AIXPRT is now available to the public! AIXPRT includes support for the Intel OpenVINO, TensorFlow, and NVIDIA TensorRT toolkits to run image-classification and object-detection workloads with the ResNet-50 and SSD-MobileNet v1networks, as well as a Wide and Deep recommender system workload with the Apache MXNet toolkit. The test reports FP32, FP16, and INT8 levels of precision.

To access AIXPRT, visit the AIXPRT download page. There, a download table displays the AIXPRT test packages. Locate the operating system and toolkit you wish to test and click the corresponding Download link. For detailed installation instructions and information on hardware and software requirements for each package, click the package’s Readme link. If you’re not sure which AIXPRT package to choose, the AIXPRT package selector tool will help to guide you through the selection process.

In addition, the Helpful Info box on AIXPRT.com contains links to a repository of AIXPRT resources, as well links to XPRT blog discussions about key AIXPRT test configuration settings such as batch size and precision.

We hope AIXPRT will prove to be a valuable tool for you, and we’re thankful for all the input we received during the preview period! If you have any questions about AIXPRT, please let us know.

Check out the other XPRTs: