• Corpus ID: 203952657

A Fast Array of Wimpy Nodes

  title={A Fast Array of Wimpy Nodes},
  author={David G. Andersen and Jason Franklin and Amar Phanishayee and Lawrence Tan and Vijay Vasudevan},
This paper introduces the FAWN—Fast Array of Wimpy Nodes—cluster architecture for providing fast, scalable, and power-efficient key-value storage. A FAWN links together a large number of tiny nodes built using embedded processors and small amounts (2–16GB) of flash memory into an ensemble capable of handling 700 queries per second per node, while consuming fewer than 6 watts of power per node. We have designed and implemented a clustered key-value storage system, FAWN-DHT, that runs atop these… 
LSM-trie: An LSM-tree-based Ultra-Large Key-Value Store for Small Data Items
LSM-trie is proposed, a KV storage system that substantially reduces metadata for locating KV items, reduces write amplification by an order of magnitude, and needs only two disk accesses with each KV read even when only less than 10% of metadata (Bloom filters) can be held in memory.
MICA: A Holistic Approach to Fast In-Memory Key-Value Storage
MICA optimizes for multi-core architectures by enabling parallel access to partitioned data, and for efficient parallel data access, MICA maps client requests directly to specific CPU cores at the server NIC level by using client-supplied information and adopts a light-weight networking stack that bypasses the kernel.
RAIL: A Case for Redundant Arrays of Inexpensive Links in Data Center Networks
RAIL, a system to ensure that in a data center networks, applications only use paths that meet their performance needs, can reduce the network cost by up to 10% for 10 Gbps networks and 44% for 40Gbps networks, without affecting the applications’ performance.
Thin servers with smart pipes: designing SoC accelerators for memcached
This work argues for an alternate architecture---Thin Servers with Smart Pipes (TSSP)---for cost-effective high-performance memcached deployment, and demonstrates the potential benefits of the TSSP architecture through an FPGA prototyping platform, and shows the potential for a 6X-16X power-performance improvement over conventional server baselines.
TBF: A memory-efficient replacement policy for flash-based caches
A new RAM-frugal cache replacement policy that approximates the least-recently-used (LRU) policy is proposed that uses two in-memory Bloom sub-filters (TBF) for maintaining the recency information and leverages an on-flash key-value store to cache objects.
Scaling Out to a Single-Node 80Gbps Memcached Server with 40Terabytes of Memory
Novel hybrid memory systems that combine conventional DRAMs and serial-attached flash to increase value store capacity to 40Terabytes with up to 200 million entries while providing access at 80Gbps are presented.
BOOM: Enabling mobile memory based low-power server DIMMs
Buffered Output On Module (BOOM) is proposed that buffers the data outputs from multiple ranks of low-frequency mobile DRAM devices, which in aggregation provide high bandwidth and achieve chipkill-correct or even stronger reliability.
SkimpyStash: RAM space skimpy key-value store on flash-based storage
SkimpyStash is a RAM space skimpy key-value store on flash-based storage, designed for high throughput, low latency server applications and provides throughputs from few 10,000s to upwards of 100,000 get-set operations/sec.
Achieving power-efficiency in clusters without distributed file system complexity
This work describes a cluster built with power-efficient node prototypes and presents experimental evaluations to demonstrate power-efficiency, and exploits cluster nodes that have the ability to operate in at least two extreme system level power states, characterized by minimum vs. maximum power consumption and performance.
FlashStore: High Throughput Persistent Key-Value Store
This work uses real-world data traces from two data center applications, namely, Xbox LIVE Primetime online multi-player game and inline storage deduplication, to drive and evaluate the design of FlashStore on traditional and low power server platforms.


Storage alternatives for mobile computers
This paper investigates three alternative storage devices for mobile computers: magnetic hard disks, flash memory disk emulators, and flash memory cards and shows that flash memory can reduce energy consumption by an order of magnitude, compared to magnetic disk, while providing good read performance and acceptable write performance.
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.
SmartSaver: Turning Flash Drive into a Disk Energy Saver for Mobile Computers
This paper proposes to make another highly desired use of the flash drive - saving disk energy by using the flashdrive as a standby buffer for caching and prefetching disk data.
Filling the Memory Access Gap: A Case for On-Chip Magnetic Storage (CMU-CS-99-174)
Results from applications show that MEMS-based storage can reduce application I/O stall times by 80{99%, with overall performance improvements ranging from 1.1 to 20 for these applications.