Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

A number of recent efforts have attempted to design accelerators for popular machine learning algorithms, such as those involving convolutional and deep neural networks (CNNs and DNNs). These algorithms typically involve a large number of multiply-accumulate (dot-product) operations. A recent project, DaDianNao, adopts a near data processing approach, where a specialized neural functional unit performs all the digital arithmetic operations and receives input weights from adjacent eDRAM banks.This work explores an in-situ processing approach, where memristor crossbar arrays not only store input weights, but are also used to perform dot-product operations in an analog manner. While the use of crossbar memory as an analog dot-product engine is well known, no prior work has designed or characterized a full-fledged accelerator based on crossbars. In particular, our work makes the following contributions: (i) We design a pipelined architecture, with some crossbars dedicated for each neural network layer, and eDRAM buffers that aggregate data between pipeline stages. (ii) We define new data encoding techniques that are amenable to analog computations and that can reduce the high overheads of analog-to-digital conversion (ADC). (iii) We define the many supporting digital components required in an analog CNN accelerator and carry out a design space exploration to identify the best balance of memristor storage/compute, ADCs, and eDRAM storage on a chip. On a suite of CNN and DNN workloads, the proposed ISAAC architecture yields improvements of 14.8×, 5.5×, and 7.5× in throughput, energy, and computational density (respectively), relative to the state-of-the-art DaDianNao architecture.

ISAAC: a convolutional neural network accelerator with in-situ analog arithmetic in crossbars

A key question for the microprocessor research and design community is whether scaling multicores will provide the performance and value needed to scale down many more technology generations. To provide a quantitative answer to this question, a comprehensive study that projects the speedup potential of future multicores and examines the underutilization of integration capacity-dark silicon-is timely and crucial.

/pdf/dark-silicon-and-the-end-of-multicore-scaling-fxij8qeoty.pdf

Dark Silicon and the End of Multicore Scaling

Since 2005, processor designers have increased core counts to exploit Moore's Law scaling, rather than focusing on single-core performance. The failure of Dennard scaling, to which the shift to multicore parts is partially a response, may soon limit multicore scaling just as single-core scaling has been curtailed. This paper models multicore scaling limits by combining device scaling, single-core scaling, and multicore scaling to measure the speedup potential for a set of parallel workloads for the next five technology generations. For device scaling, we use both the ITRS projections and a set of more conservative device scaling parameters. To model single-core scaling, we combine measurements from over 150 processors to derive Pareto-optimal frontiers for area/performance and power/performance. Finally, to model multicore scaling, we build a detailed performance model of upper-bound performance and lower-bound core power. The multicore designs we study include single-threaded CPU-like and massively threaded GPU-like multicore chip organizations with symmetric, asymmetric, dynamic, and composed topologies. The study shows that regardless of chip organization and topology, multicore scaling is power limited to a degree not widely appreciated by the computing community. Even at 22 nm (just one year from now), 21% of a fixed-size chip must be powered off, and at 8 nm, this number grows to more than 50%. Through 2024, only 7.9x average speedup is possible across commonly used parallel workloads, leaving a nearly 24-fold gap from a target of doubled performance per generation.

/pdf/dark-silicon-and-the-end-of-multicore-scaling-goghgnng6o.pdf

Dark silicon and the end of multicore scaling

Augmenting Amdahl's law with a corollary for multicore hardware makes it relevant to future generations of chips with multiple processor cores. Obtaining optimal multicore performance will require further research in both extracting more parallelism and making sequential cores faster.

/pdf/amdahl-s-law-in-the-multicore-era-zghnwm3c06.pdf

Amdahl's Law in the Multicore Era

Memory scaling is in jeopardy as charge storage and sensing mechanisms become less reliable for prevalent memory technologies, such as DRAM. In contrast, phase change memory (PCM) storage relies on scalable current and thermal mechanisms. To exploit PCM's scalability as a DRAM alternative, PCM must be architected to address relatively long latencies, high energy writes, and finite endurance.We propose, crafted from a fundamental understanding of PCM technology parameters, area-neutral architectural enhancements that address these limitations and make PCM competitive with DRAM. A baseline PCM system is 1.6x slower and requires 2.2x more energy than a DRAM system. Buffer reorganizations reduce this delay and energy gap to 1.2x and 1.0x, using narrow rows to mitigate write energy and multiple rows to improve locality and write coalescing. Partial writes enhance memory endurance, providing 5.6 years of lifetime. Process scaling will further reduce PCM energy costs and improve endurance.

/pdf/architecting-phase-change-memory-as-a-scalable-dram-3v0ii94s6h.pdf

Architecting phase change memory as a scalable dram alternative

Modern computer systems have been built around the assumption that persistent storage is accessed via a slow, block-based interface. However, new byte-addressable, persistent memory technologies such as phase change memory (PCM) offer fast, fine-grained access to persistent storage.In this paper, we present a file system and a hardware architecture that are designed around the properties of persistent, byteaddressable memory. Our file system, BPFS, uses a new technique called short-circuit shadow paging to provide atomic, fine-grained updates to persistent storage. As a result, BPFS provides strong reliability guarantees and offers better performance than traditional file systems, even when both are run on top of byte-addressable, persistent memory. Our hardware architecture enforces atomicity and ordering guarantees required by BPFS while still providing the performance benefits of the L1 and L2 caches.Since these memory technologies are not yet widely available, we evaluate BPFS on DRAM against NTFS on both a RAM disk and a traditional disk. Then, we use microarchitectural simulations to estimate the performance of BPFS on PCM. Despite providing strong safety and consistency guarantees, BPFS on DRAM is typically twice as fast as NTFS on a RAM disk and 4-10 times faster than NTFS on disk. We also show that BPFS on PCM should be significantly faster than a traditional disk-based file system.

/pdf/better-i-o-through-byte-addressable-persistent-memory-2wrl2k9p36.pdf

Better I/O through byte-addressable, persistent memory

Efficiently utilizing off-chip DRAM bandwidth is a critical issuein designing cost-effective, high-performance chip multiprocessors(CMPs). Conventional memory controllers deliver relativelylow performance in part because they often employ fixed,rigid access scheduling policies designed for average-case applicationbehavior. As a result, they cannot learn and optimizethe long-term performance impact of their scheduling decisions,and cannot adapt their scheduling policies to dynamic workloadbehavior.We propose a new, self-optimizing memory controller designthat operates using the principles of reinforcement learning (RL)to overcome these limitations. Our RL-based memory controllerobserves the system state and estimates the long-term performanceimpact of each action it can take. In this way, the controllerlearns to optimize its scheduling policy on the fly to maximizelong-term performance. Our results show that an RL-basedmemory controller improves the performance of a set of parallelapplications run on a 4-core CMP by 19% on average (upto 33%), and it improves DRAM bandwidth utilization by 22%compared to a state-of-the-art controller.

Self-Optimizing Memory Controllers: A Reinforcement Learning Approach

Phase-change may enable continued scaling of main memories, but PCM has higher access latencies, incurs higher power costs, and wears out more quickly than DRAM. This article discusses how to mitigate these limitations through buffer sizing, row caching, write reduction, and wear leveling, to make PCM a viable dream alternative for scalable main memories.

/pdf/phase-change-technology-and-the-future-of-main-memory-3k4rbjr317.pdf

Phase-Change Technology and the Future of Main Memory

Architects use cycle-by-cycle simulation to evaluate design choices and understand tradeoffs and interactions among design parameters. Efficiently exploring exponential-size design spaces with many interacting parameters remains an open problem: the sheer number of experiments renders detailed simulation intractable. We attack this problem via an automated approach that builds accurate, confident predictive design-space models. We simulate sampled points, using the results to teach our models the function describing relationships among design parameters. The models produce highly accurate performance estimates for other points in the space, can be queried to predict performance impacts of architectural changes, and are very fast compared to simulation, enabling efficient discovery of tradeoffs among parameters in different regions. We validate our approach via sensitivity studies on memory hierarchy and CPU design spaces: our models generally predict IPC with only 1-2% error and reduce required simulation by two orders of magnitude. We also show the efficacy of our technique for exploring chip multiprocessor (CMP) design spaces: when trained on a 1% sample drawn from a CMP design space with 250K points and up to 55x performance swings among different system configurations, our models predict performance with only 4-5% error on average. Our approach combines with techniques to reduce time per simulation, achieving net time savings of three-four orders of magnitude.

/pdf/efficiently-exploring-architectural-design-spaces-via-4vbwqffuyl.pdf

Engin Ipek

Papers

Architecting phase change memory as a scalable dram alternative

Better I/O through byte-addressable, persistent memory

Self-Optimizing Memory Controllers: A Reinforcement Learning Approach

Phase-Change Technology and the Future of Main Memory

Efficiently exploring architectural design spaces via predictive modeling