Challenges and opportunities for efficient computing with FAWN

  title={Challenges and opportunities for efficient computing with FAWN},
  author={Vijay Vasudevan and David G. Andersen and Michael Kaminsky and Jason Franklin and Michael A. Kozuch and Iulian Moraru and Padmanabhan Pillai and Lawrence Tan},
  journal={ACM SIGOPS Oper. Syst. Rev.},
This paper presents the architecture and motivation for a clusterbased, many-core computing architecture for energy-efficient, dataintensive computing. FAWN, a Fast Array of Wimpy Nodes, consists of a large number of slower but efficient nodes coupled with low-power storage. We present the computing trends that motivate a FAWN-like approach, for CPU, memory, and storage. We follow with a set of microbenchmarks to explore under what workloads these FAWN nodes perform well (or perform poorly… 

Figures and Tables from this paper

Survey on energy-efficient hard drive disks
  • Evans Oliech, X. Ruan
  • Computer Science
    2017 International Conference on Computing, Networking and Communications (ICNC)
  • 2017
This paper presents some of proposed and implemented ideas to make storage systems energy efficient with minor performance degradation and analyzes the concepts.
An Energy-Efficient and Reliable Storage Mechanism for Data-Intensive Academic Archive Systems
An evaluation function called E3SaRC (Economic Evaluation of Energy Saving with Reliability Constraint), which considers the cost of hardware failure when applying energy-saving schemes, and an adaptive write cache mechanism called CacheRAID, which addresses the issue of system reliability.
CacheRAID: An Efficient Adaptive Write Cache Policy to Conserve RAID Disk Array Energy
The experimental results show that Cache RAID storage system can conserved 50%~70% of the power consumption compared to the conventional software RAID system.
Compact Filter Structures for Fast Data Partitioning
FilterKV is presented, a data management scheme for faster online data partitioning of key-value (KV) pair data that reduces the amount of data shuffled over the network by: (a) moving KV pairs quickly off the network to storage, and (b) using an extremely compact representation to represent each KV pair in the communication occurring over thenetwork.
A study of suitable fault tolerance frameworks for an energy-efficient storage system
This study proposes a general energy-efficient framework that can be integrated different fault tolerance mechanisms and experimental results show that the capability of the proposed framework with different erasure code algorithms can avoid data loss.
Modeling System Power Consumption Considering DVFS and Thermal Effect
This paper model the power consumption of a single server with resource utilization considering two factors, DVFS and thermal effect, and proposes a new system architecture based on this model.
Energy-Agility: A New Grid-centric Metric for Evaluating System Performance
Energy-eciency has long been considered a \rst class" metric for evaluating computer system performance. A high energy-eciency translates into more work done per unit of energy, which increases the


FAWN: a fast array of wimpy nodes
The key contributions of this paper are the principles of the FAWN architecture and the design and implementation of FAWN-KV--a consistent, replicated, highly available, and high-performance key-value storage system built on a FAWN prototype.
Gordon: using flash memory to build fast, power-efficient clusters for data-intensive applications
The paper presents an exhaustive analysis of the design space of Gordon systems, focusing on the trade-offs between power, energy, and performance that Gordon must make, and describes a novel flash translation layer tailored to data intensive workloads and large flash storage arrays.
Low Power Amdahl Blades for Data Intensive Computing
The emergence of Solid State Disk (SSD) technology, poses the challenge of building a credible equivalent to a GrayWulf system with a similar IO performance, but with considerably lower power consumption, than an SSD-based system, while maintaining the same cost and power consumption.
Building Energy-Efficient Systems for Sequential I/O Workloads
This paper designs an SSD-based system for highly energy-efficient sequential I/O, and demonstrates that by trading latency for power, this system can achieve similar energy efficiency across a variety of systems, from embedded-class to desktop and server-class systems.
Active Disks for Large-Scale Data Processing
This work proposes using an active disk storage device that combines on-drive processing and memory with software downloadability to allow disks to execute application-level functions directly at the device.
Optimizing Power Consumption in Large Scale Storage Systems
This paper presents a simple idea that allows the storage system to turn off a large fraction of its disks, without incurring unacceptable performance penalties, and presents a simulator that allows us to explore this parameter space.
Evaluation of Existing Architectures in IRAM Systems
This work examined both execution time analyses of existing microprocessors and system simulation of hypothetical processors to determine whether existing microarchitectures can tap the potential performance advantages of IRAM systems.
Energy-efficient sorting using solid state disks
Understanding and Designing New Server Architectures for Emerging Warehouse-Computing Environments
A new solution that incorporates volume non-server-class components in novel packaging solutions, with memory sharing and flash-based disk caching, has promise, with a 2X improvement on average in performance-per-dollar for the benchmark suite.
Low-power amdahl-balanced blades for data intensive computing
This paper proposes an alternative architecture comprising large number of so-called Amdahl blades that combine energy-efficient CPUs with solid state disks to increase sequential read I/O throughput by an order of magnitude while keeping power consumption constant.