Academic
Publications
Practical and low-overhead masking of failures of TCP-based servers

Practical and low-overhead masking of failures of TCP-based servers,10.1145/1534909.1534911,ACM Transactions on Computer Systems,Dmitrii Zagorodnov,Ke

Practical and low-overhead masking of failures of TCP-based servers   (Citations: 4)
BibTex | RIS | RefWorks Download
This article describes an architecture that allows a replicated service to survive crashes without breaking its TCP connections. Our approach does not require modifications to the TCP protocol, to the operating system on the server, or to any of the software running on the clients. Furthermore, it runs on commodity hardware. We compare two implementations of this architecture (one based on primary/backup replication and another based on message logging) focusing on scalability, failover time, and application transparency. We evaluate three types of services: a file server, a Web server, and a multimedia streaming server. Our experiments suggest that the approach incurs low over- head on throughput, scales well as the number of clients increases, and allows recovery of the service in near-optimal time.
Journal: ACM Transactions on Computer Systems - TOCS , vol. 27, no. 2, pp. 1-39, 2009
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...Splitting is similar to [16] and using the M-TCP (Migr migrate TCP connections from on However, connection migration in splitting an HTTP request and operating in split mode...

    Bharat S. Rawalet al. Splitting HTTP requests on two servers

    • ...Several of these schemes require primary and backup servers to process requests in parallel, e.g., FT-TCP (hot backup) [24] and ST-TCP [14]...
    • ...FT-TCP [24] accomplishes TCP failover by logging (persistently storing) every packet in a TCP connection on a primary server to a backup server...

    Wyatt Lloydet al. Coercing clients into facilitating failover for object delivery

    • ...TCP connection is different from connections, processes or Web sessions connections; or masking failures in TCP-ba example, in migratory TCP (M-TCP) connection is migrated between serve involvement; in process migration [11] process is transferred between machines; session handoff [2], a proxy is used to migra in a mobile environment; in TCP splicing [ TCP connections are established for each fault-tolerant TCP (FT-TCP) [16], a T continues ...

    Bharat S. Rawalet al. Mini Web Server Clusters for HTTP Request Splitting

    • ...This is commonly referred to as the output commit problem [4] and the introduced latency in transmitting the reply is called the output commit stall [5]...
    • ...Despite the availability of solutions for overcoming certain types of nondeterminism, such as random numbers [13], a recent paper [5] showed that identifying most sources can be cumbersome...
    • ...Furthermore, synchronizing such events between the primary and the backup machines often requires modifications to the source code of the server application, which may not be always available [5]...
    • ...after replying the client), which is a fundamental limitation of any fault-tolerant system [5]...
    • ...Extensions of this work [11], [5] investigate exploiting PWD by means of wrapping the TCP stack and logging not only packets and socket read calls, but also other system calls that may form sources of non-determinism...
    • ...It is reportedly necessary to modify most applications in order to identify and synchronize all non-determinism [5]...

    B. Gerofiet al. A Multi-core Approach to Providing Fault Tolerance for Non-determinist...

Sort by: