The CAT theorem and performance of transactional distributed systems

2016 ◽  
Author(s):  
Shegufta Bakht Ahsan ◽  
Indranil Gupta
Keyword(s):  
Distributed Systems ◽  
And Performance

Energy- and performance-aware scheduling of tasks on parallel and distributed systems

2012 ◽  
Vol 8 (4) ◽  
pp. 1-37 ◽  
Author(s):  
Hafiz Fahad Sheikh ◽  
Hengxing Tan ◽  
Ishfaq Ahmad ◽  
Sanjay Ranka ◽  
Phanisekhar Bv
Keyword(s):  
Distributed Systems ◽  
And Performance

UNIVERSAL ROUTING AND PERFORMANCE ASSURANCE FOR DISTRIBUTED NETWORKS

2007 ◽  
Vol 08 (01) ◽  
pp. 1-28
Author(s):  
KEVIN F. CHEN ◽  
EDWIN H.-M. SHA
Keyword(s):  
Distributed Systems ◽  
Large Scale ◽  
Distributed Networks ◽  
Distributed Network ◽  
Graph Layout ◽  
Universal Network ◽  
Wire Delay ◽  
And Performance ◽  
The Way

We show that universal routing can be achieved with low overhead in distributed networks. The validity of our results rests on a new network called the fat-stack. We show that from a routing perspective the fat-stack is efficient and is suitable for use as a baseline distributed network and as a crucial benchmark architecture for evaluating the performance of specific distributed networks. We show that the fat-stack is efficient by proving it is universal. A requirement for the fat-stack to be universal is that link capacities double up the levels of the network. We use methods developed in the areas of VLSI and processor interconnect for much of our analysis. We then show how to scale the fat-stack from a VLSI graph layout to a large-scale distributed topology and how the network can be an effective benchmark architecture. Our universality proofs show that a fat-stack of area Θ(A) can simulate any competing network of area A with [Formula: see text] overhead independently of wire delay. The universality result implies that the fat-stack of a given size is nearly the best routing network of that size. The fat-stack is also the minimal universal network for an [Formula: see text] overhead in terms of number of links. Actual simulations show that the fat-stack outperforms a mesh-based distributed network of comparable hardware usage. Our work helps explain why some deployed networks function in the way they do in terms of routing. It also provides an exemplary network of proven efficiency and scalability for building new distributed systems.

scholarly journals Single / multi TCP and a fairness issue

2017 ◽  
pp. 08-16
Author(s):  
Dunia Shwehdy ◽  
Fatma Badish ◽  
Nadia Badish
Keyword(s):  
Distributed Systems ◽  
High Performance ◽  
Network Performance ◽  
Parallel Tcp ◽  
And Performance

In distributed systems domain a good connection is a structural quest for messages transfer between clusters in today's high performance applications. Initiating a single TCP used in networks, than it drops as a bundle of TCP's called multiple or parallel TCP. This drop wasn't quite successful it also created some ugly fouling , for this reason some studies concluded that in general resultant a multi TCP is not much differ of a single TCP ,but others refined that the multi socket TCP achieves better throughput and performance than the single socket approach [1]. The applications that require good network performance often use multi socket TCP streams and TCP modifications to improve the effectiveness of TCP .but if the network bottleneck is fully utilized, this approach boosts throughput by unfairly stealing bandwidth from competing TCP stream. This dilemma driven to some opinions see that improve the effectiveness of TCP is much easier than to while maintaining fairness [2]. in this paper we submit our own destination to compare between single-socket TCP and multi-socket TCP to conclusiveness the argumentativeness according about any one is the best, in performance and throughput ,we also discussed a fairness issue in both ,and finally we defined an algorithm for enhance multi TCP's performance.

Sign in / Sign up

Export Citation Format

Share Document