*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: Towards Building Scalable, Reliable and Secure Flash-based Storage Systems
Jian Huang
http://www.cc.gatech.edu/~jhuang95/
Ph.D. student
School of Computer Science
College of Computing
Georgia Institute of Technology
Date: Monday, November 7th, 2016
Time: 11:30am - 1:30pm
Location: KACB 2100
Committee:
-------------------------------------------------
Dr. Moinuddin K. Qureshi (Advisor, School of Electrical and Computer Engineering, Georgia Tech)
Dr. Umakishore Ramachandran (School of Computer Science, Georgia Tech)
Dr. Taesoo Kim (School of Computer Science, Georgia Tech)
Dr. Steven Swanson (Department of Computer Science and Engineering, University of California, San Diego)
Dr. James Mickens (Department of Computer Science, Harvard University)
Dr. Anirudh Badam (Systems Research Group, Microsoft Research, Redmond)
Abstract:
--------------------------------------------------
The longstanding goals of storage systems design have been to provide simple abstractions
for high-level applications to efficiently access data while ensuring the data durability and
security on physical device. The traditional storage system, which was designed for slow hard
disk with block interface, does not fit for the recent hardware trends such as the faster flash
memory with more flexible interfaces. The gap between the storage system software and
hardware device not only causes resource inefficiency and sub-optimal performance, but also
result in uncertainty of data security and durability on storage.
To bridge this gap, we rethink the system design with a holistic approach from system level to
device level for SSDs. First, we present FlashMap, a holistic SSD architecture that combines
memory, storage and device-level indirections into a unified layer. By combining these layers,
FlashMap reduces critical-path latency for data access and improves DRAM caching efficiency
significantly for SSDs with large capacity.
Second, we propose FlashBlox, an SSD virtualization system that utilizes flash parallelism to
improve performance isolation between virtual SSDs by assigning them on dedicated channels
and dies. It allows the wear of different channels and dies to diverge at fine time granularity in
favor of isolation and adjusts that imbalance at a more coarse time granularities in a principled
manner. FlashBlox reduces the tail latency of storage operations dramatically while achieving
uniform lifetime for SSDs.
Third, we present FlashGuard, a new SSD which has the capability of defending against
encryption ransomware by leveraging the intrinsic characteristics of SSDs. FlashGuard can
restore user files encrypted by malware, while adding minimal overhead to normal storage
operations.
