A Case for Buffer Servers
نویسندگان
چکیده
Faster networks and cheaper storage have brought us to a point where I/O caching servers have an important role in the design of scalable, high-performance file systems. These intermediary I/O servers — or buffer servers — can be deployed at strategic points in the network, interposed between clients and data sources such as standard file servers, Internet data servers, and tertiary storage. Their purpose is to provide a fast and incrementally scalable I/O service throughout the network while reducing and smoothing demands on shared data servers and the network backbone. This position paper outlines a case for caching buffer servers and addresses some of the key technical challenges in the design of a buffer service. We also describe the role of buffer servers in the Trapeze project, which uses gigabit networks as a vehicle for high-speed network I/O.
منابع مشابه
Queue with Heterogeneous Server Under Resequencing Constraint
In this paper, a markovian queue with two types of servers and buffer space is considered. The resequencing constraint is imposed according to which the customers leave the system in the same order in which they entered it. For finite buffer queue, the steady state queue size distribution has been obtained.?
متن کاملOn Two-Echelon Multi-Server Queue with Balking and Limited Intermediate Buffer
In this paper we study two echelon multi-server tandom queueing systems where customers arrive according to a poisson process with two different rates. The service rates at both echelons are independent of each other. The service times of customers is assumed to be completed in two stages. The service times at each stage are exponentially distributed. At the first stage, the customers may balk ...
متن کاملBuffer Size Requirements for Delay Sensitive Traffic Considering Discrete Effects and Service-Latency in ATM Switches
Various approaches to buffer size and management for output buffering in ATM switches supporting delay sensitive traffic are reviewed. Discrete worst case arrival and service functions are presented. Using this format, bounds are developed for buffer size under zero cell loss for leaky bucket constrained sources. Tight bounds are developed for the case of discrete arrival functions with fluid s...
متن کاملVacation model for Markov machine repair problem with two heterogeneous unreliable servers and threshold recovery
Markov model of multi-component machining system comprising two unreliable heterogeneous servers and mixed type of standby support has been studied. The repair job of broken down machines is done on the basis of bi-level threshold policy for the activation of the servers. The server returns back to render repair job when the pre-specified workload of failed machines is build up. The first (seco...
متن کاملA case for network-centric buffer cache organization
The emergence of clustered and networked storage architecture gives rise to a new type of servers which act as data conduits over the network for remotely stored data. These servers, which we call pass-through servers, are mainly responsible for passing the data through them without interpreting it in any way. In this paper, we put forward a scheme of network-centric buffer cache management. Th...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1999