Millions Of Servers Research Articles

A Hamilton Path Embedding Algorithm on DCell

DCell has been proposed for enormous data centers as a server centric interconnection which can support millions of servers with high network capacity and provide good fault tolerance by only using commodity switches. In this paper, we study an attractive algorithm of finding Hamilton paths in dimensional. The time complexity of this algorithm is , where is the node number of DCellk .

Review of the Web Metrics Model

Today, the web is growing at a fast pace. The Internet is a network connecting millions of servers and billions of people around the world. Life without the Internet is nothing. However, very little research has been conducted on the size of the Internet. This paper has been written to review the most popular web metrics and most importantly to examine how it can play a role in improving information on the web and also the quality of that information. Here, we present a basic comparison between web analytics and web metrics. We have gone deep into the subject of web metrics, discussing its different paradigms.

DPillar: Dual-port server interconnection network for large scale data centers

To meet the huge demands of computation power and storage space, a future data center may have to include up to millions of servers. The conventional hierarchical tree-based data center network architecture faces several challenges in scaling a data center to that size. Previous research effort has shown that a server-centric architecture, where servers are not only computation and storage workstations but also intermediate nodes relaying traffic for other servers, performs well in scaling a data center to a huge number of servers. This paper presents a server-centric data center network called DPillar, whose topology is inspired by the classic butterfly network. DPillar provides several nice properties and achieves the balance between topological scalability, network performance, and cost efficiency, which make it suitable for building large scale future data centers. Using only commodity hardware, a DPillar network can easily accommodate millions of servers. The structure of a DPillar network is symmetric so that any network bottleneck is eliminated at the architectural level. With each server having only two ports, DPillar is able to provide the bandwidth to support communication intensive distributed applications. This paper studies the interconnection features of DPillar, how to compute routes in DPillar, and how to forward packets in DPillar. Extensive simulation experiments have been performed to evaluate the performance of DPillar. The results show that DPillar performs well even in the presence of a large number of server and switch failures.

DAW: A Distributed Antiworm System

A worm automatically replicates itself across networks and may infect millions of servers in a short period of time. It is conceivable that the cyberterrorists may use a widespread worm to cause major disruption to the Internet economy. Much recent research concentrates on propagation models and early warning, but the defense against worms is largely an open problem. We propose a distributed antiworm architecture (DAW) that automatically slows down or even halts the worm propagation within an Internet service provider (ISP) network. New defense techniques are developed based on the behavioral difference between normal hosts and worm-infected hosts. Particularly, a worm-infected host has a much higher connection-failure rate when it randomly scans the Internet. This property allows DAW to set the worms apart from the normal hosts. We propose a temporal rate-limit algorithm and a spatial rate-limit algorithm, which makes the speed of worm propagation configurable by the parameters of the defense system. The effectiveness of the new techniques is evaluated analytically and by simulations.

Efficient performance models for layered server systems with replicated servers and parallel behaviour

Capacity planning for large computer systems may require very large performance models, which are difficult or slow to solve. Layered queueing models solved by mean value analysis can be scaled to dozens of servers and hundreds of service classes, with large class populations, but this may not be enough. A common feature of planning models for large systems is structural repetition expressed through replicated subsystems, which can provide both scalability and reliability, and this replication can be exploited to scale the solution technique. A model has recently been described for symmetrically replicated layered servers, and their integration into the system, with a mean-value solution approximation. However, parallelism is often combined with replication; high-availability systems use parallel data-update operations on redundant replicas, to enhance reliability, and grid systems use parallel computations for scalability. This work extends the replicated layered server model to systems with parallel execution paths. Different servers may be replicated to different degrees, with different relationships between them. The solution time is insensitive to the number of replicas of each replicated server, so systems with thousands or even millions of servers can be modelled efficiently.

Discontinuities driven by a billion connected machines

We have witnessed an explosion of Internet use that has tremendous implications for our industry, particularly on the test requirements put upon us. At Intel we have a vision. We see a world in which we're marching toward a billion connected computers being served by high-performance networks and millions of servers. On those servers we see billions of transactions worth trillions of dollars, conducted on a daily and a yearly basis. This connected world is changing everything-everything that we do and everything that we need to deliver as an industry.

The Microsoft database research group

Microsoft’s strategic interest in the database field dates from 1993 and the efforts of David Vaskevitch, who is now the Microsoft Vice President in charge of the database and transaction processing product development groups. David’s vision was that the world would need millions of servers, and that this presented a wonderful opportunity to a company like Microsoft that sells software in high volume and at low prices. Database systems played an important role in Vaskevitch’s vision, and, indeed, in Microsoft’s current product plans. David began looking for premier database and transaction processing people in late 1993. The scope of Vaskevitch’s efforts included a desire for Microsoft to establish a database research group. Rick Rashid, Microsoft Research Vice President, collaborated with Vaskevitch in recruiting David Lomet from Digital’s Cambridge Research Lab to initiate the Microsoft Database Research Group. Lomet joined Microsoft Research in January of 1995. Hence, Microsoft’s Database Research Group is now a little over three and a half years old. One person does not a group make. Recruiting efforts continued. Surajit Chaudhuri, a researcher from HP Labs joined the Database Group in February of 1996. Paul Larson, a professor from the University of Waterloo joined in May of that year. Vivek Narasayya was initially an intern as a graduate student from the University of Washington in the summer of 1996, officially joining the group in April of 1997. Roger Barga, the newest member of the group and a new Oregon Graduate Institute Ph.D., joined in December, 1997.