Remote Memory Backing Storage for Distributed Virtual Memory Operating Systems (Thesis)
James Griffoen · Purdue e-Pubs (Purdue University System) · 1992
The virtual memory abstraction aids in the design and implementation of portable applications.Virtual memory allows applications to execute in an arbitrarily large memory space, independent from the physical memory size of the underlying machine.Conventional virtual memory operaling systems use magnetic disks for backing storage.lvlagnetlc disks provide high data transfer rates, large storage capacity, and the abHlty to randomly access data, making them an appealing backing storage medium.However, the average seek Lime of a magnetic disk is several orders of magnitude slower than memory access times.Recent advances in CPU speeds, network bandwidth, and memory sizes have made new types of backing storage with improved performance and greater functionality feasible.This thesis investigates a new model for virtual memory in which dedicated, large-memory machines provide backing storage to virtual memory systems executing on a set of client machines in a distributed environment.Dedicated memory se1'vers provide clients with a large, shared memory resource.Each memory server machine is capable of supporting heterogeneous client machines executing a wide variety of operating systems.Clients that exceed the capacity of their local memory access remote memory servers across a high-speed network to obtain addltional storage space.Clients use a highly efficient, special •This paper was originally published as: James Griffioen, Remote Memory Backing Storage for Distributed Virtual Memory Operating Systems, PI1D thesis, Department of Computer Science, Purdue University, West Lafayette, IN, August 1991.This technical report was created in response to a sudden (and somewhat delayed) barrage of requests for the original thesis.Please use the original thesis citation when citing this work.purpose, reliable 1 data streaming, network architecture independent communication protocol to transfer data to and from the memory server.To reduce the delay associated with accessing remote memory, memory servers use efficient algorithms and data structures to retrieve data, on average, in constant time.In addition to providing a highly-efficient backing store, the model allows data sharing between clients and improves file system performance by offioading the me server.Tills thesis also describes the design and implementation of a prototype system.Measurements obtained from the prototype implementation clearly demonstrate the viability of systems based on the model.The prototype shows that remote memory systems offer performance competitive with, and in some cases better than, existing virtual memory systems.Moreover, rapid advances in network bandwidth, CPU speeds, and memory sizes make the model an attractive ba.<>is for the design of future distributed systems.