138 research outputs found
BFT: The time is now
No full textData centers strive to provide reliable access to the data and services that they host. This reliable access requires the hosted data and services hosted by the data center to be both consistent and available. Byzantine fault tolerance (BFT) replication offers the promise of services that are consistent and available despite arbitrary failures by a bounded number of servers and an unbounded number of clients
Making Byzantine Fault Tolerant Systems Tolerate Byzantine Faults
No full textThis paper argues for a new approach to building Byzantine fault tolerant replication systems. We observe that although recently developed BFT state machine replication protocols are quite fast, they don't tolerate Byzantine faults very well: a single faulty client or server is capable of rendering PBFT, Q/U, HQ, and Zyzzyva virtually unusable. In this paper, we (1) demonstrate that existing protocols are dangerously fragile, (2) define a set of principles for constructing BFT services that remain useful even when Byzantine faults occur, and (3) apply these principles to construct a new protocol, Aardvark. Aardvark can achieve peak performance within 40% of that of the best existing protocol in our tests and provide a significant fraction of that performance when up to f servers and any number of clients are faulty. We observe useful throughputs between 11706 and 38667 requests per second for a broad range of injected faults
Note on Singular Observation
No full textThe author (Jørgen Dahlin), highly experienced dentist within practicing, university research and teaching within his specialization, recollects after 44 years his personal sensed observation of a hypotetized information transfer between two closely family related persons, during the death process of one of the persons. The line of sight distance of the information transfer was 8 km
Hierarchical Cache Consistency in WAN (Extended Abstract)
No full textJian Yin, Lorenzo Alvisi, Mike Dahlin, Calvin Lin Department of Computer Sciences University of Texas at Austin Abstract This paper explores ways to provide strong consistency for Internet applications scaling to millions of clients. We make four contributions. First, we identify the ways in which specific characteristics of data-access workloads affect the scalability of cache consistency algorithms. Second, we define two primitive mechanisms, split and join for growing and shrinking hierarchies. We show how these primitives can be implemented with a simple mechanism already present in a protocol for strong consistency that we have previously proposed. Third, we describe and evaluate policies for using split and join to address the fault tolerance and performance challenges of hierarchies. Finally, we compare various algorithms for maintaining strong consistency in a range of hierarchy configurations. We evaluate our algorithms using simulations. 1 Introduction To prevent the ra..
FlightPath: obedience vs. choice in cooperative services
No full textWe present FlightPath, a novel peer-to-peer streaming application that provides a highly reliable data stream to a dynamic set of peers. We demonstrate that FlightPath reduces jitter compared to previous works by several orders of magnitude. Furthermore, FlightPath uses a number of run-time adaptations to maintain low jitter despite 10% of the population behaving maliciously and the remaining peers acting selfishly. At the core of FlightPath's success are approximate equilibria. These equilibria allow us to design incentives to limit selfish behavior rigorously, yet they provide sufficient flexibility to build practical systems. We show how to use an Ɛ-Nash equilibrium, instead of a strict Nash, to engineer a live streaming system that uses bandwidth efficiently, absorbs flash crowds, adapts to sudden peer departures, handles churn, and tolerates malicious activity
Recommended from our members
Transparent replication
No full textIncreasing user expectations and demands have caused the evolution of
web services away from single-server systems and toward distributed systems
for their ability to provide improved throughput, improved availability and reduced
response times. However, for a service to run on a distributed system,
each running instance must be able to access data that are shared among the
instances. Although existing off-the-shelf replication systems - e.g. distributed
file systems [52, 61, 32, 38, 41], replicated databases [64, 75], distributed hash
tables [58, 59, 63, 34], etc. - simplify access to shared data by exporting wellresearched
interfaces, their implementations are typically not engineered for
the unique environments presented by many web services. For example, replication
systems that require synchronization across multiple nodes to handle
modified data [38, 12] or systems that require all nodes to keep a copy of all
data [64, 75] may not be practical for use in such services.
Although the problem of general replication is not possible to solve [11,
62, 33] we focus our study on a class of single-writer services that we denote
Information Dissemination Services that form a restrictive but important set
of web services.
Our research makes two key contributions. First, we show that for a
class of single-writer services that we denote Information Dissemation Services
TRIP replicates dynamic data in a manner that is nearly transparent to the
service. We (1) develop a novel dual-channel replication algorithm for TRIP
that utilizes spare network background traffic to speculatively replicate data in
a safe, non-interfering fashion, (2) show how to integrate safe speculative replication
with mechanisms that use invalidates to provide consistency, and (3)
demonstrate how our combination of consistency and safe speculative replication
allows us to provide near-ideal consistency, performance, and availability
for Information Dissemination Services.
Second, we show that the core principles behind building TRIP can be
extended to build a new replication framework and more general replication
toolkit. In particular, we show that it is possible to extend our dual-queue
mechanisms developed for TRIP to a multi-writer environment where nodes
can synchronize multiple incoming streams of data and consistency information.
Our extension allows providing various forms of consistency for arbitrary
topologies - two key properties provided by the PRACTI [6] (Partial Replication,
Arbitrary Consistency, Topology Independence) architecture.Computer Scienc
Recommended from our members
SAR: semantic-aware replication
No full texttextThis dissertation presents a replication framework that facilitates semantic-aware data
replication (SAR) in wide area networks (WANs). WAN data replication is fundamentally
difficult. As a result, generic replication algorithms must make compromises among
Consistency, Availability, Response time, and Partition resilience (CARP) when used in
WANs. This dissertation seeks to design algorithms based on specific semantics of the
shared data sets (e.g. data properties, workload characteristics, and update patterns) to
achieve the optimized CARP trade-offs. Integrating a set of semantic-aware algorithms using
distributed objects to form the SAR framework, we implement a practically important
e-commerce application, the distributed TPC-W benchmark. Our prototype evaluations
show significant improvements on system availability and response time while preserving
the consistency guarantees desired by the TPC-W benchmark. The primary focus of the
dissertation is on the development of the SAR framework. Within the framework, contributions
include (a) exploiting application semantics using the object-oriented approach,
(b) employing a hybrid method that integrates a number of novel replication algorithms to
make an important class of applications work, (c) proposing a novel replication algorithm
for the multi-writer/multi-reader replication scenario with a high access locality, and (d)
outlining a general purpose replication library that uses semantic-aware objects for building
other distributed applications in WANs.Computer Scienc
Recommended from our members
URA : a universal data replication architecture
No full texttextData replication is a key building block for large-scale distributed systems to improve availability, performance, and scalability. Because there is a fundamental trade-off between performance and consistency as well as between availability and consistency, systems must make trade-offs among these factors based on the demands and technologies of their target environments and workloads. Unfortunately, existing replication protocols and mechanisms are intrinsically entangled with specific policy assumptions. Therefore, to accommodate new trade-offs for new policy requirements, developers have to either build a new replication system from scratch or modify existing mechanisms. This dissertation presents a universal data replication architecture (URA) that cleanly separates mechanism and policy and supports Partial Replication (PR), Any Consistency (AC), and Topology Independence (TI) simultaneously. Our architecture yields two significant advantages. First, by providing a single set of mechanisms that capture the common underlying abstractions for data replication, URA can serve as a common substrate for building and deploying new replication systems. It therefore can significantly reduce the effort required to construct or modify a replication system. Second, by providing a set of general and flexible mechanisms independent of any specific policy, URA enables better trade-offs than any current system can provide. In particular, URA can simultaneously provide the three PRACTI properties while any existing system can provide at most two of them. Our experimental results and case-study systems confirm that universal data replication architecture is a way to build better replication systems and a better way to build replication systems.Computer Scienc
Recommended from our members
Volume lease: a scalable cache consistency framework
No full texttextComputer Scienc
Happy diamond anniversary JMS! A decade analysis of the Journal of Management Studies
No full textThe Journal of Management Studies, founded in 1963, is celebrating its 60 th year. Clark et al. (2014) conducted a bibliometric analysis for its 50 th anniversary assessing whether the journal had maintained its leading international ranking and sustained its mission to serve as a broad-based management outlet. In this review, we build on and extend their findings by examining trends in the journal over the past decade (2012–22). We present a broader analysis of JMS by exploring its unique identity within the management journal ecosystem and examining its scope and breadth in terms of topics, methods, and author demographics to document JMS's evolution, impact, reach, and accessibility. We develop a new bibliometric framework that employs a mix of qualitative and quantitative analyses (including regression, text, and language analysis) to cover a broad range of considerations for a journal and its stakeholders. In so doing, we contribute to the bibliometric and review research areas by proposing new metrics (related to diversity, equity, and inclusion) and analysis tools to assess the relative position of an academic journal. Employing this framework, we conclude that JMS has retained and enhanced its position as a leading, cutting-edge general management journal.</p
- …
