Hung-Sheng Chang

Marching-Based Wear-Leveling for PCM-Based Storage Systems

Improving the performance of storage systems without losing the reliability and sanity/integrity of file systems is a major issue in storage system designs. In contrast to existing storage architectures, we consider a PCM-based storage architecture to enhance the reliability of storage systems. In PCM-based storage systems, the major challenge falls on how to prevent the frequently updated (meta)data from wearing out their residing PCM cells without excessively searching and moving metadata around the PCM space and without extensively updating the index structures of file systems. In this work, we propose an adaptive wear-leveling mechanism to prevent any PCM cell from being worn out prematurely by selecting appropriate data for swapping with constant search/sort cost. Meanwhile, the concept of indirect pointers is designed in the proposed mechanism to swap data without any modification to the file systems indexes. Experiments were conducted based on well-known benchmarks and realistic workloads to evaluate the effectiveness of the proposed design, for which the results are encouraging.

A PCM translation layer for integrated memory and storage management

Phase change memory (PCM) is known for its potentials as main memory and as storage. In contrast to the past work, this work presents a PCM translation layer that considers how the main memory is used by the operating system together with the usage patterns of storage by trading their performance and reliability. In particular, a joint management scheme is proposed to improve the capability of PCM as both main memory and storage so as to enhance the performance of the entire system. The endurance issue of PCM for main memory is resolved by utilizing the potentially large capacity of the PCM storage space with limited modifications to the existing operating system implementations. Moreover, three commands are proposed to help operating system engineers to take advantage of PCM as main memory and storage by reducing I/O overheads and speeding up both the initialization and termination of program executions. The experimental results show that the performance of PCM as both main memory and storage can be significantly improved with reasonable lifetime, while the system overhead is very limited under coarse-grained wear leveling.

A Light-Weighted Software-Controlled Cache for PCM-based Main Memory Systems

The replacement of DRAM with non-volatile memory relies on solutions to resolve the wear leveling and slow write problems. Different from the past work in compiler-assisted optimization or joint DRAM-PCM management strategies, we explore a light-weighted software-controlled DRAM cache design for the non-volatile-memory-based main memory. The run-time overheads in the management of the DRAM cache is minimized by utilizing the information from a miss of the translation lookaside buffer (TLB) or the cache. Experiments were conducted based on a series of the well-known benchmarks to evaluate the effectiveness of the proposed design, for which the results are very encouraging.

How to improve the space utilization of dedup-based PCM storage devices?

There is a growing demand to introduce more and more intelligence to storage devices in recent years, especially with the rapid increasing of hardware computing power. This paper targets on essential design issues in space utilization for dedup-based non-volatile phase-change memory (PCM). We explore the adoption of data duplication techniques to reduce potential data duplicates over PCM storage devices to provide more storage space than the physical storage space does. Among various data deduplication techniques, variable-sized chunking is considered in less cost-effective PCM-based storage devices because variable-sized chunking has better data deduplication capability than fixed-sized chunking. However, in a typical system architecture, data are written or updated in the fixed management units (e.g., LBAs). Thus, to ultimately improve the space utilization of PCM-based storage device, the technical problem falls on (1) how to map fixed-sized LBAs to variable-sized chunks and (2) how to efficiently manage (i.e., allocated and deallocate) free PCM storage space for variable-sized chunks. In this work, we propose a free space manager, called container-based space manager, to resolve the above two issues by exploiting the fact that (1) a storage system initially has more free space to relax the complexity on space management and (2) the space optimization of a storage system can grow with the time when it contains more and more data. The proposed design is evaluated over popular benchmarks, for which we have very encouraging results.

A disturbance-aware sub-block design to improve reliability of 3D MLC flash memory

The reliability problem of modern flash-memory chips quickly deteriorates because of the nature of MLC chips. Although the vertical stacking of storage cells in 3D flash-memory chips dramatically increases the bit density, compared to 2D chips, it also results in severe disturbance problems. In this work, we propose to create sub-blocks in the MTD layer by considering different program disturb resistance patterns, without any hardware cost. In particular, a disturbance-aware sub-block design is proposed to utilize the hotness information of data to further improve the reliability of 3D MLC flash memory by smartly choosing sub-blocks to use when extra information is available from the flash translation layer. The proposed design was evaluated by a series of experiments over traces from SNIA. It was shown that the disturb errors could be reduced by at least 50% under DFTL and BL, i.e., a page-level FTL and a block-level FTL design, compared to the conventional MLC programming designs.

Pattern-aware write-back strategy to minimize energy consumption of PCM-based storage systems

Phase-change memory (PCM) is a potential candidate to replace flash memory in the storage design of mobile computing systems, but its energy consumption becomes a challenging issue when the multi-level cell (MLC) technology is adopted to reduce its unit cost. In contrast to the existing works that focus on reducing energy consumption at the hardware/device level, we propose an energy-aware memory management design with a pattern-aware write-back strategy in the software/OS level. It selects and writes proper data from main memory to PCM device with minimized energy consumption with considering the data bit pattern, data access locality, and characteristics of PCM. The experiments were conducted based on the workloads collected from representative benchmarks to evaluate the capability of the proposed design, and the results are very encouraging.