ASIC clouds

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.

[1]  Steven Swanson,et al.  Conservation cores: reducing the energy of mature computations , 2010, ASPLOS XV.

[2]  Michael Bedford Taylor,et al.  A Landscape of the New Dark Silicon Design Regime , 2013, IEEE Micro.

[3]  Steven Swanson,et al.  GreenDroid: A mobile application processor for a future of dark silicon , 2010, 2010 IEEE Hot Chips 22 Symposium (HCS).

[4]  ZhangLu,et al.  Extreme Datacenter Specialization for Planet-Scale Computing , 2018 .

[5]  Michael Bedford Taylor,et al.  Is dark silicon useful? Harnessing the four horsemen of the coming dark silicon apocalypse , 2012, DAC Design Automation Conference 2012.

[6]  Moein Khazraee,et al.  Specializing a Planet's Computation: ASIC Clouds , 2017, IEEE Micro.

[7]  Vikram Bhatt,et al.  The GreenDroid Mobile Application Processor: An Architecture for Silicon's Dark Future , 2011, IEEE Micro.

[8]  L. V. Gutierrez,et al.  ASIC Clouds: Specializing the Datacenter , 2016, 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA).

[9]  Lu Zhang,et al.  Extreme Datacenter Specialization for Planet-Scale Computing: ASIC Clouds , 2018, OPSR.

[10]  David A. Patterson,et al.  In-datacenter performance analysis of a tensor processing unit , 2017, 2017 ACM/IEEE 44th Annual International Symposium on Computer Architecture (ISCA).

[11]  Steven Swanson,et al.  Efficient complex operators for irregular codes , 2011, 2011 IEEE 17th International Symposium on High Performance Computer Architecture.

[12]  Lu Zhang,et al.  Moonwalk: NRE Optimization in ASIC Clouds , 2017, ASPLOS.

[13]  Muhammad Shafique,et al.  The EDA challenges in the dark silicon era , 2014, 2014 51st ACM/EDAC/IEEE Design Automation Conference (DAC).