Implementation of a virtual SW platform for the SSD-based memory sub-system evaluation and exploration
keywords DIGITAL SYSTEM DESIGN TEST AND VERIFICATION, SOFTWARE ENGINEERING, VIRTUAL MACHINE
Reference persons STEFANO DI CARLO, PAOLO ERNESTO PRINETTO
External reference persons INDACO Marco (PhD Candidate), GALFANO Salvatore (PhD Candidate)
Research Groups TESTGROUP - TESTGROUP
Thesis type EXPERIMENTAL
Currently solid state drives (SSDs) based on NAND flash memories are predominant in the consumer electronic market.
However, SSD-based memory sub-system has quickly become determinant of the performances, power consumption and reliability of modern computing system.
Many optimizations can be done at different levels of the overall memory architecture. Namely, a memory sub-system is composed of several levels (from the bottom to the top): the NAND memory chips, its controller, drivers, file system.
So, evaluating how a given memory sub-system can impact on the application performances is a very important challenge to be faced.
So far, several virtual platforms are raised to give the capabilities to the SW/HW engineer to design the overall hardware/software architecture in order to assess the overall system performances (e.g., Open Virtual Platform (OVP)), but currently no support is provided for SSD drives. On the contrary, the only available SSD drive simulator, called Disksim is used to evaluate just modern storage subsystem performances but it doesn’t offer a support to emulate a fully working system to assess the overall system-level performances.
The goal of this thesis is to design and implement an emulation environment combining Disksim and Open Virtual Platform tools. In particular an ARM-based evaluation platform will be designed.
The candidate will acquire a basic knowledge of system level design methodology, skills on using the main emulation tools and a good knowledge of SSD-based memory devices.
Required skills Programming Languages: C/C++
Notes Number of required Students: 1 or 2
Deadline 01/03/2015 PROPONI LA TUA CANDIDATURA