GPU and FPGA-based clouds have already demonstrated the promise of accelerating computing-intensive workloads with greatly improved power and performance.In this paper, we examine the design of ASIC Clouds, which are purpose-built datacenters comprised of large arrays of ASIC accelerators, whose purpose is to optimize the total cost of ownership (TCO) of large, high-volume chronic computations, which are becoming increasingly common as more and more services are built around the Cloud model. On the surface, the creation of ASIC clouds may seem highly improbable due to high NREs and the inflexibility of ASICs. Surprisingly, however, large-scale ASIC Clouds have already been deployed by a large number of commercial entities, to implement the distributed Bitcoin cryptocurrency system.We begin with a case study of Bitcoin mining ASIC Clouds, which are perhaps the largest ASIC Clouds to date. From there, we design three more ASIC Clouds, including a YouTubestyle video transcoding ASIC Cloud, a Litecoin ASIC Cloud, and a Convolutional Neural Network ASIC Cloud and show 2-3 orders of magnitude better TCO versus CPU and GPU.Among our contributions, we present a methodology that given an accelerator design, derives Pareto-optimal ASIC Cloud Servers, by extracting data from place-and-routed circuits and computational fluid dynamic simulations, and then employing clever but brute-force search to find the best jointlyoptimized ASIC, DRAM subsystem, motherboard, power delivery system, cooling system, operating voltage, and case design. Moreover, we show how data center parameters determine which of the many Pareto-optimal points is TCOoptimal. Finally we examine when it makes sense to build an ASIC Cloud, and examine the impact of ASIC NRE.The first two authors contributed equally. To Appear in ISCA 2016.1 See http://blockchain.info to see real-time ledger updates.

We explore a novel approach for building DNN training clusters using commodity optical devices. Our proposal, called TOPOOPT, co-optimizes the distributed training process across three dimensions: computation, communication, and network topology. TOPOOPT uses a novel alternating optimization technique and a group theory-inspired algorithm to find the best network topology and routing plan, together with parallelization strategy, for distributed DNN training. To motivate our proposal, we measure the communication patterns of distributed DNN workloads at a large online service provider. Experiments with a 12-node prototype demonstrate the feasibility of TOPOOPT. Simulations on real distributed training models show that, compared to similar-cost Fat-tree interconnects, TOPOOPT reduces DNN training time by up to 3×.

Planet-scale applications are driving the exponential growth of the cloud, and datacenter specialization is the key enabler of this trend, providing order of magnitudes improvements in cost-effectiveness and energy-efficiency. While exascale computing remains a goal for supercomputing, specialized datacenters have emerged and have demonstrated beyond-exascale performance and efficiency in specific domains. This paper generalizes the applications, design methodology, and deployment challenges of the most extreme form of specialized datacenter: ASIC Clouds. It analyzes two game-changing, real-world ASIC Clouds-Bitcoin Cryptocurrency Clouds and Tensor Processing Clouds-discuss their incentives, the empowering technologies and how they benefit from the specialized ASICs. Their business models, architectures and deployment methods are useful for envisioning future potential ASIC Clouds and forecasting how they will transform computing, the economy and society.

GPU and FPGA-based clouds have already demonstrated the promise of accelerating computing-intensive workloads with greatly improved power and performance.In this paper, we examine the design of ASIC Clouds, which are purpose-built datacenters comprised of large arrays of ASIC accelerators, whose purpose is to optimize the total cost of ownership (TCO) of large, high-volume chronic computations, which are becoming increasingly common as more and more services are built around the Cloud model. On the surface, the creation of ASIC clouds may seem highly improbable due to high NREs and the inflexibility of ASICs. Surprisingly, however, large-scale ASIC Clouds have already been deployed by a large number of commercial entities, to implement the distributed Bitcoin cryptocurrency system.We begin with a case study of Bitcoin mining ASIC Clouds, which are perhaps the largest ASIC Clouds to date. From there, we design three more ASIC Clouds, including a YouTubestyle video transcoding ASIC Cloud, a Litecoin ASIC Cloud, and a Convolutional Neural Network ASIC Cloud and show 2-3 orders of magnitude better TCO versus CPU and GPU.Among our contributions, we present a methodology that given an accelerator design, derives Pareto-optimal ASIC Cloud Servers, by extracting data from place-and-routed circuits and computational fluid dynamic simulations, and then employing clever but brute-force search to find the best jointlyoptimized ASIC, DRAM subsystem, motherboard, power delivery system, cooling system, operating voltage, and case design. Moreover, we show how data center parameters determine which of the many Pareto-optimal points is TCOoptimal. Finally we examine when it makes sense to build an ASIC Cloud, and examine the impact of ASIC NRE.The first two authors contributed equally. To Appear in ISCA 2016.1 See http://blockchain.info to see real-time ledger updates.

Planet-scale applications are driving the exponential growth of the Cloud, and datacenter specialization is the key enabler of this trend. GPU- and FPGA-based clouds have already been deployed to accelerate compute-intensive workloads. ASIC-based clouds are a natural evolution as cloud services expand across the planet. ASIC Clouds are purpose-built datacenters comprised of large arrays of ASIC accelerators that optimize the total cost of ownership (TCO) of large, high-volume scale-out computations. On the surface, ASIC Clouds may seem improbable due to high NREs and ASIC inflexibility, but large-scale ASIC Clouds have already been deployed for the Bitcoin cryptocurrency system. This paper distills lessons from these Bitcoin ASIC Clouds and applies them to other large scale workloads such as YouTube-style video-transcoding and Deep Learning, showing superior TCO versus CPU and GPU. It derives Pareto-optimal ASIC Cloud servers based on accelerator properties, by jointly optimizing ASIC architecture, DRAM, motherboard, power delivery, cooling, and operating voltage. Finally, the authors examine the impact of ASIC NRE and when it makes sense to build an ASIC Cloud.