A distributed lock manager on fault tolerant MPP
نویسندگان
چکیده
The Distributed L.ock Manager (DLM) is the cornerstone of running the Oracle Parallel Server on a MPP UNIX system. The DLM grants resource locks in NULL, READ, WRITE, and other levels. The DLM is typically implemented either with distributed message passing or in global shared memory, depending on the architecture. Reflective memory on the Encore Infinity 90 architecture permits a shared memory style DLM. Each node runs a separate copy of UNIX so failure of one node is isolated and DLM recovery can cleanup state. A node can later rejoin the Oracle session after no& repair.
منابع مشابه
Stabilis: A Case Study in Writing Fault-Tolerant Distributed Applications Using Persistent Objects
This paper presents Stabilis, a fault-tolerant object-oriented distributed database management system that has been written as an exercise in persistent programming. Sta-bilis is implemented on top of Arjuna, an object-oriented programming system that provides the basic mechanisms for fault tolerance and distribution. The computational model used by Arjuna is based upon the concept of using ato...
متن کاملUsing Virtualization to Validate Fa Ult-tolerant Distributed Systems
Asynchronous events and complex system state distributed across independent nodes make exposure and diagnosis of flaws in distributed systems a challenge. The difficulties are exacerbated when the goal is to validate fault tolerance mechanisms that are activated only by the occurrence of errors, which are, by nature, rare. Validation of fault tolerance mechanisms is often done by injecting faul...
متن کاملFault Tolerant Parallel Image Generation on a Workstation Network
Image generation for computer movies is a good candidate application for parallelisation. This application was used as a starting point to design a fault tolerant distributed computing environment aimed to run parallel applications. The paper rst describes the context of this work, then it presents the requirements that the environment should meet. The paper then describes the use of the Distri...
متن کاملVerifying a Design Pattern for the Fault-Tolerant Execution of Parallel Programs∗
2 The protocol 4 2.1 The setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.2 The parallel program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.3 The client’s protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.3.1 The read operation . . . . . . . . . . . . ....
متن کامل