-
- News
- Books
Featured Books
- pcb007 Magazine
Latest Issues
Current IssueMaterial Matters
Materials management is nuanced, multifaceted, and requires a holistic systems approach for business success. When building high mix, low volume, and high technology, managing materials and overall cost containment are even greater challenges.
Additive Manufacturing
In this month’s issue, we explore additive manufacturing technology for the PCB fabricator: where it stands today, the true benefits, and where it seems to be headed.
The Growing Industry
In this issue of PCB007 Magazine, we talk with leading economic experts, advocacy specialists in Washington, D.C., and PCB company leadership to get a well-rounded picture of what’s happening in the industry today. Don’t miss it.
- Articles
- Columns
Search Console
- Links
- Events
||| MENU - pcb007 Magazine
Faster Big-Data Analysis
October 31, 2017 | MITEstimated reading time: 5 minutes
We live in the age of big data, but most of that data is “sparse.” Imagine, for instance, a massive table that mapped all of Amazon’s customers against all of its products, with a “1” for each product a given customer bought and a “0” otherwise. The table would be mostly zeroes.
With sparse data, analytic algorithms end up doing a lot of addition and multiplication by zero, which is wasted computation. Programmers get around this by writing custom code to avoid zero entries, but that code is complex, and it generally applies only to a narrow range of problems.
At the Association for Computing Machinery’s Conference on Systems, Programming, Languages and Applications: Software for Humanity (SPLASH), researchers from MIT, the French Alternative Energies and Atomic Energy Commission, and Adobe Research recently presented a new system that automatically produces code optimized for sparse data.
That code offers a 100-fold speedup over existing, non-optimized software packages. And its performance is comparable to that of meticulously hand-optimized code for specific sparse-data operations, while requiring far less work on the programmer’s part.
The system is called Taco, for tensor algebra compiler. In computer-science parlance, a data structure like the Amazon table is called a “matrix,” and a tensor is just a higher-dimensional analogue of a matrix. If that Amazon table also mapped customers and products against the customers’ product ratings on the Amazon site and the words used in their product reviews, the result would be a four-dimensional tensor.
“Sparse representations have been there for more than 60 years,” says Saman Amarasinghe, an MIT professor of electrical engineering and computer science (EECS) and senior author on the new paper. “But nobody knew how to generate code for them automatically. People figured out a few very specific operations — sparse matrix-vector multiply, sparse matrix-vector multiply plus a vector, sparse matrix-matrix multiply, sparse matrix-matrix-matrix multiply. The biggest contribution we make is the ability to generate code for any tensor-algebra expression when the matrices are sparse.”
Joining Amarasinghe on the paper are first author Fredrik Kjolstad, an MIT graduate student in EECS; Stephen Chou, also a graduate student in EECS; David Lugato of the French Alternative Energies and Atomic Energy Commission; and Shoaib Kamil of Adobe Research.
Custom kernels
In recent years, the mathematical manipulation of tensors — tensor algebra — has become crucial to not only big-data analysis but machine learning, too. And it’s been a staple of scientific research since Einstein’s time.
Traditionally, to handle tensor algebra, mathematics software has decomposed tensor operations into their constituent parts. So, for instance, if a computation required two tensors to be multiplied and then added to a third, the software would run its standard tensor multiplication routine on the first two tensors, store the result, and then run its standard tensor addition routine.
In the age of big data, however, this approach is too time-consuming. For efficient operation on massive data sets, Kjolstad explains, every sequence of tensor operations requires its own “kernel,” or computational template.
“If you do it in one kernel, you can do it all at once, and you can make it go faster, instead of having to put the output in memory and then read it back in so that you can add it to something else,” Kjolstad says. “You can just do it in the same loop.”
Computer science researchers have developed kernels for some of the tensor operations most common in machine learning and big-data analytics, such as those enumerated by Amarasinghe. But the number of possible kernels is infinite: The kernel for adding together three tensors, for instance, is different from the kernel for adding together four, and the kernel for adding three three-dimensional tensors is different from the kernel for adding three four-dimensional tensors.
Many tensor operations involve multiplying an entry from one tensor with one from another. If either entry is zero, so is their product, and programs for manipulating large, sparse matrices can waste a huge amount of time adding and multiplying zeroes.
Hand-optimized code for sparse tensors identifies zero entries and streamlines operations involving them — either carrying forward the nonzero entries in additions or omitting multiplications entirely. This makes tensor manipulations much faster, but it requires the programmer to do a lot more work.
The code for multiplying two matrices — a simple type of tensor, with only two dimensions, like a table — might, for instance, take 12 lines if the matrix is full (meaning that none of the entries can be omitted). But if the matrix is sparse, the same operation can require 100 lines of code or more, to track omissions and elisions.
Enter Taco
Taco adds all that extra code automatically. The programmer simply specifies the size of a tensor, whether it’s full or sparse, and the location of the file from which it should import its values. For any given operation on two tensors, Taco builds a hierarchical map that indicates, first, which paired entries from both tensors are nonzero and, then, which entries from each tensor are paired with zeroes. All pairs of zeroes it simply discards.
Taco also uses an efficient indexing scheme to store only the nonzero values of sparse tensors. With zero entries included, a publicly released tensor from Amazon, which maps customer ID numbers against purchases and descriptive terms culled from reviews, takes up 107 exabytes of data, or roughly 10 times the estimated storage capacity of all of Google’s servers. But using the Taco compression scheme, it takes up only 13 gigabytes — small enough to fit on a smartphone.
“Many research groups over the last two decades have attempted to solve the compiler-optimization and code-generation problem for sparse-matrix computations but made little progress,” says Saday Sadayappan, a professor of computer science and engineering at Ohio State University, who was not involved in the research. “The recent developments from Fred and Saman represent a fundamental breakthrough on this long-standing open problem.”
“Their compiler now enables application developers to specify very complex sparse matrix or tensor computations in a very easy and convenient high-level notation, from which the compiler automatically generates very efficient code,” he continues. “For several sparse computations, the generated code from the compiler has been shown to be comparable or better than painstakingly developed manual implementations. This has the potential to be a real game-changer. It is one of the most exciting advances in recent times in the area of compiler optimization.”
Suggested Items
Altair Names Semantic Partners Channel Partner for the EMEA Region
06/28/2024 | AltairAltair, a global leader in computational intelligence, announced its collaboration with Semantic Partners as a channel partner for the EMEA region.
Wild River Technology's ISI-USB4 Now Supports USB 4.0 Version 2 and DisplayPort 2.1
06/27/2024 | Wild River TechnologyWild River Technology (WRT), the leading supplier of signal integrity measurement and optimization test fixtures for high-speed channels at data rates of up to 25G announced today the availability of a new inter-symbol interference (ISI) loss modeling capability. The ISI-USB4 Advanced Loss Modeling Platform Test Card Set delivers comprehensive support for demanding 25G PAM4 designs.
Altair Names Devoteam Channel Partner for the EMEA Region
06/26/2024 | AltairAltair, a global leader in computational intelligence, is thrilled to welcome Devoteam as a new channel partner for the EMEA region. Within the partnership, Devoteam will offer Altair’s leading data analytics and artificial intelligence (AI) solutions to its customers throughout the EMEA region.
IDC Reveals Automotive Semiconductor Market Outlook and Discusses Technological Advancements
06/26/2024 | IDCA recent IDC report, Worldwide Automotive Semiconductor Market Forecast, 2024 categorizes automotive chips into distinct segments: analog chips, logic chips, microprocessors, memory chips, discrete devices, optoelectronic devices, and sensors.
Book Excerpt: The Printed Circuit Assembler’s Guide to... Factory Analytics—Chapter 1
06/26/2024 | I-Connect007 Editorial TeamDespite being a widely discussed topic since the advent of Industry 4.0 initiatives, the implementation and utilization of analytics tools by electronics manufacturers has been slow, with an adoption rate significantly lower than anticipated. Most of this stagnation can be attributed to the fact that getting data from the factory floor simply isn’t as straightforward as it should be.