• Publications
  • Influence
Squirrel: a decentralized peer-to-peer web cache
TLDR
This paper presents a decentralized, peer-to-peer web cache called Squirrel, which exhibits performance comparable to a centralized web cache in terms of hit ratio, bandwidth usage and latency. Expand
  • 441
  • 27
  • PDF
Anticipatory scheduling: a disk scheduling framework to overcome deceptive idleness in synchronous I/O
TLDR
We propose the anticipatory disk scheduling framework to solve this problem in a simple, general and transparent way, based on the non-work-conserving scheduling discipline. Expand
  • 270
  • 21
  • PDF
Practical, transparent operating system support for superpages
TLDR
We propose an effective superpage management system for FreeBSD on the Alpha CPU, and evaluate it on real workloads and benchmarks. Expand
  • 136
  • 7
  • PDF
Practical, transparent operating system support for superpages
Most general-purpose processors provide support for memory pages of large sizes, called superpages. Superpages enable each entry in the translation lookaside buffer (TLB) to map a large physicalExpand
  • 80
  • 4
A Resource Management Framework for Predictable Quality of Service in Web Servers
This paper presents a resource management framework for providing predictable quality of service (QoS) in Web servers. The framework allows Web server and proxy operators to ensure a probabilisticExpand
  • 20
  • PDF
Druschel the effect of deceptive idleness on disk schedulers
TLDR
We solve deceptive idleness by designing a transparent, non-work-conserving scheduling framework for various scheduling policies. Expand
  • 9
Advanced memory management and disk scheduling techniques for general-purpose operating systems
TLDR
Adaptive memory management is a technique that informs applications of the severity of memory pressure via a metric that quantifies the cost of using memory. Expand
  • 3
A performance study of snoopy and directory based cache-coherence protocols
—There are two dominant schemes of cache-coherence protocols in use today, namely snoopy-and directory-based. In snoopy-based systems, all coherence transactions are broadcast and therefore seen byExpand
  • 1
Application-assisted physical memor y management for general-purpose operating systems
Manysoftware applicationscan, in principle, trademain memoryconsumption for other resources. For instance, garbagecollectedlanguageruntimescantradecollection overheadfor heap size,Expand
Application-assist ed physical memor y management
Many software applications can, in principle, trade main memory consumption for other resources. For instance, garbage collected language runtimes can trade collection overhead for heap size, andExpand