Reducing initial latency in a multimedia storage system

Author(s):  
E. Chang ◽  
H. Garcia-Molina
2008 ◽  
Vol 1129 ◽  
Author(s):  
Dong Jin Jung ◽  
Kinam Kim

AbstractWe demonstrate that ferroelectric memory is very eligible to become a non-volatile cache solution, in particular, in a multimedia storage system such as solid-state disk. It could provide benefits both of performance and of reliability. In performance, a FRAM cache allows us to rid overhead of power-off recovery. Random WRITE performance has been improved by 250%. In assertion of endurance, we investigate acceleration factors to evaluate cycle-to-failure of the ferroelectric memory both in device-level and in capacitor-level. What has been found is that ferroelectric memory cells have 6.0×1014 of the cycle-to-failure at the operational condition of 85 o C and 2.0V. This cycle-to-failure is well above lifetime READ/WRITE cycles of 9.5×1013 in such system. From 2-dimensional stress simulation, it has also been concluded that the number of dummy cells plays a critical role in qualifying the high temperature life tests.


Author(s):  
Phillip K.C. Tse

Multimedia storage systems store data objects and receive streams of requests from the multimedia server. When a client wishes to display an object, it sends a new object request for the multimedia object to the multimedia server as shown in Figure 15.1. The multimedia server checks to see if this new stream can be accepted. If accepted, the server sends a data request to the storage system to retrieve the first data stripe. The storage system returns the data stripe to the server. The server then encapsulates the data stripe as data packets and sends the data packets to the client. The client extracts the data stripe from the data packets. Afterwards, the server sends data requests periodically to the storage system. Each of these data requests has a deadline associated with it. If the request cannot be served before the deadline, the client program does not have any more data to display. The stream thus will be suspended or aborted. Therefore, every request of a stream, except the first one, must be served within the deadline to ensure continuity of the stream. Before we consider the scheduling methods for request streams in the next chapter, we describe the feasibility to accept concurrent streams in this chapter.


1995 ◽  
Vol 41 (1) ◽  
pp. 53-64 ◽  
Author(s):  
Yen-Jen Oyang ◽  
Chun-Hung Wen ◽  
Chih-Yuan Cheng ◽  
Meng-Huang Lee ◽  
Jian-Tian Li

Sign in / Sign up

Export Citation Format

Share Document