In recent years, Intel has been talking about its Cascade Lake servers with DL Boost (also called VNNI, Vector Neural Net Instructions). These new features are a subset of the AVX-512 and are intended to specifically accelerate processor performance in artificial intelligence applications. Historically, many AI applications have favored GPUs over processors. The GPU architecture is much better suited to graphics processors than processors. Processors offer many more thread-based execution resources, but even today's multicore processors are overwhelmed by the parallelism available in a high-end graphics processor core.
Anandtech did you compare the performance of Cascade Lake, the Epyc 7601 (soon outperformed by the AMD Rome 7nm processors, but still today AMD's main server core), and a RTX Titan. The article, written by the excellent Johan De Gelas, discusses different types of neural networks beyond CNN networks (convolutional neural networks), which are generally compared, and explains how a key element of the strategy Intel is competing with Nvidia in workloads where GPUs are not as powerful. or can not yet meet the emerging market needs due to memory capacity constraints (GPUs still can not match CPUs here), the use of 'light' artificial intelligence models does not not requiring long periods of workout, or artificial intelligence models that rely on statistical models of non-neural networks.
Growing revenue from data centers is a critical part of Intel's global strategy for artificial intelligence and machine learning. Nvidia, meanwhile, is keen to protect a market in which it currently has virtually no competition. Intel's Artificial Intelligence strategy is broad and encompasses many products, from Movidius and Nervana to DL Boost on Xeon, to the next GPU Xe range. Nvidia seeks to show that GPUs can be used to handle artificial intelligence calculations in a wider range of workloads. Intel incorporates new artificial intelligence features into its existing products, uses new hardware that, it is hoped, will impact the market, and attempts to create its first serious GPU to challenge the work done by AMD and Nvidia in the consumer market.
Anandtech's benchmarks show, overall, that the gap between Intel and Nvidia remains wide, even with DL Boost. This graph of a recurrent neural network test used a "Long Short Term Memory (LSTM)" network as a neural network. A type of RNN, LSTM "selectively remembers" patterns over a period of time. "Anandtech also used three different configurations to test it – Tensorflow ready to use with conda, a Tensorflow optimized for Intel with PyPi, and optimized version of Tensorflow from the source using Bazel, using the latest version of Tensorflow.
This pair of images captures the relative scale between the processors as well as the comparison with the RTX Titan. Ready-to-use performance was quite poor under AMD, although it improved with the optimized code. Intel's performance skyrocketed like a rocket when the source-optimized version was tested, but even the source-optimized version did not fit the performance of Titan RTX very well. De Gelas notes: "Secondly, we were pretty surprised that our Titan RTX is less than three times faster than our dual Xeon setup," which explains how these comparisons are done in the larger article.
DL Boost is not enough to narrow the gap between Intel and Nvidia, but in all fairness, this has probably never been supposed to be. Intel's goal is to improve AI performance enough on Xeon to make the execution of these plausible workloads on servers that will be mainly used for other purposes, or when creating models of artificial intelligence that do not meet the constraints of D & C A modern graphics processor. The long-term goal of the company is to compete in the AI market with a range of equipment, not just Xeons. Since Xe is not quite ready yet, competing in the HPC space means competing with Xeon for the moment.
For those of you who are wondering about AMD, AMD does not really talk about performing artificial intelligence workloads on Epyc processors, but is focused on its RocM initiative to run CUDA code under OpenCL. AMD does not talk much about this side of its business, but Nvidia dominates the GPU application market for AI and HPC. AMD and Intel both want a piece of space. At the moment, both seem to be fighting for one.
Intel held its first-quarter 2019 teleconference yesterday. The company has revealed a lot of important information on how it expects the rest of 2019 to unfold. The news is globally mixed.
First, Intel does not expect its CPU shortage to end in the second quarter as originally planned. Bob Swan, CEO of Intel had this to say:
Our supply constraints have had a disruptive impact on our customers and our ecosystem. We have never asked for any constraints on the growth of our customers. We have increased our ability to improve our position in the second half, but the product line will remain a challenge in the third quarter as our teams align the available offer with customer demand.
This announcement implies that AMD will have more opportunities to compete, at least in the lower segments of desktop and laptop segments. Intel has focused its availability on high-end products to keep them moving, but AMD has launched new processors for Chromebooks and has gained market share in the mobile sector since early 2019.
Second, Intel says its 10nm ramp is going pretty well. Here is Swan again:
[O]Your confidence in the 10 nm nanometer also improves. In addition to improving the manufacturing speed I described earlier, we expect to qualify our first 10-nanometer volume product, Icelake, this quarter and increase our volume targets by 10 nanometers for the year.
Intel plans to increase to 10nm initially in the second quarter, but the introduction of the volume is still scheduled for the holidays, 2019. Nothing in Swan's remarks refutes or discusses the recent road map leak, which predicts no Intel 10nm desktop processor before at least 2021-2022. It is possible that Intel ignores 10 nm on the desktop and waits 7 nm before launching new parts.
Finally, Intel does not expect to see the short-term recovery of the data center predicted. After initially suggesting in Q4 that a recovery would begin in the second half of the year, Intel pushed back that date and lowered its own forecasts. The turnover is now expected to be $ 69 billion, down 3% from the previous year, a decrease of $ 2.5 billion from the previous estimate of # 39; Intel. Swan:
Our conversations with customers and partners in our PC and data-centric businesses over the last two months have highlighted several trends. The decline in memory prices has intensified. The data center inventory and digestion capacity we described in January is above our expectations, and the headwinds in China have multiplied, which has led to a more cautious IT spending environment. And yet, these same conversations with customers reinforce our confidence in improving demand in the second half of the year. So we re-evaluated our expectations for the 19's in terms of the challenges we face.
In summary, a difficult environment for Chipzilla. The 10-nm ramp continues, but leaks suggest the product will not matter much to Intel's bottom line. The data center environment is difficult and the processor shortages will continue in the third quarter. We'll have to wait to see what kind of benefit AMD can take advantage of, but even Swan acknowledged that Intel would be facing a more competitive environment in the second half of 2019, with future launches of servers and customers.