• ETRI Journal
• /
• v.25 no.5
• /
• pp.275-287
• /
• 2003

We present a high performance cache structure with a hardware prefetching mechanism that enhances exploitation of spatial and temporal locality. Temporal locality is exploited by selectively moving small blocks into the direct-mapped cache after monitoring their activity in the spatial buffer. Spatial locality is enhanced by intelligently prefetching a neighboring block when a spatial buffer hit occurs. We show that the prefetch operation is highly accurate: over 90% of all prefetches generated are for blocks that are subsequently accessed. Our results show that the system enables the cache size to be reduced by a factor of four to eight relative to a conventional direct-mapped cache while maintaining similar performance.

• Journal of IKEEE
• /
• v.18 no.3
• /
• pp.392-397
• /
• 2014

In this paper, we propose a memory operation system architecture for memory latency penalty reduction in SIMT architecture based stream processor. The proposed architecture applied non-blocking cache architecture to reduce cache miss penalty generated by blocking cache architecture. We verified that the proposed memory operation architecture improve the performance of the stream processor by comparing processing performances of various algorithms. We measured the performance improvement rate that was improved in accordance with the ratio of memory instruction in each algorithm. As a result, we confirmed that the performance of stream processor improves up to minimum 8.2% and maximum 46.5%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.3
• /
• pp.873-889
• /
• 2014

IP address lookup is very crucial in performance of routers. Several works have been done on prefix caching to enhance the performance of IP address lookup. Since a prefix represents a range of IP addresses, a prefix cache shows better performance than an IP address cache. However, not every prefix is cacheable in itself. In a prefix cache it causes false hit to cache a non-leaf prefix because there is possibly the longer matching prefix in the routing table. Prefix expansion techniques such as complete prefix tree expansion (CPTE) make it possible to cache the non-leaf prefixes as the expanded forms, but it is hard to manage the expanded prefixes. The expanded prefixes sometimes incur a great deal of update overhead in a routing table. We propose a bitmap-based prefix cache (BMCache) to provide low update overhead as well as low cache miss ratio. The proposed scheme does not have any expanded prefixes in the routing table, but it can expand a non-leaf prefix using a bitmap on caching time. The trace-driven simulation shows that BMCache has very low miss ratio in spite of its low update overhead compared to other schemes.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.11
• /
• pp.31-38
• /
• 2015

Nowadays, large size flash memory drives with more than a couple of hundreds of gigabytes are common. This paper presents an efficient cache management scheme of flash translation layer, called TPC-FTL, for large size flash memory drives. Since flash drives of large size usually contain large size RAM, we can enhance the performance of page mapping cache by using more RAM for the cache. But if the size exceeds a threshold, the existing schemes are impractical for real devices, because the time for cache manipulation becomes too long. TPC-FTL manages the cache in translation page unit, not in logical page number unit used in existing schemes. Since a translation page covers a large number of logical page numbers (for example, 512 for 2KB size page), the number of cache elements can be reduced up to a practical level. A performance evaluation shows that average response time, an important performance measure, is better than existing schemes via the effect of utilizing spacial locality in addition to temporal locality.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.4
• /
• pp.1-8
• /
• 2011

Traditional transactional memory systems are no longer able to guarantee the performance of diverse applications with overflowed transactions since there is the drawback that tracking the data for logging is difficult. Especially, this mechanism has a disadvantage of increasing communication delay for sustaining the state which is required to detect the conflict on the overflowed transactions from the first level cache in the transactional memory systems. To address this point, we have focused on the cache architecture of the systems to reduce the overhead caused by overflows and cache misses. In this paper, we present Supportive Cache which reduces additional overhead during transactions. Supportive Cache performs a parallel look-up with L1 private cache and uses the same replacement policy as L1 private cache. We evaluate the performance of the proposed design by comparing LogTM-SE with and without Supportive Cache. The simulation results show that our system improves the performance by 37% on average, compared to the original LogTM-SE which uses the same hardware resource.

• Journal of KIISE:Databases
• /
• v.32 no.1
• /
• pp.12-23
• /
• 2005

In the past decade, advances in speed of commodity CPUs have iu out-paced advances in memory latency Main-memory access is therefore increasingly a performance bottleneck for many computer applications, including database systems. To reduce memory access latency, cache memory incorporated in the memory subsystem. but cache memories can reduce the memory latency only when the requested data is found in the cache. This mainly depends on the memory access pattern of the application. At this point, previous research has shown that B+ trees perform much faster than T-trees because B+ trees are more cache conscious than T-trees, and also proposed 'Cache Sensitive B+trees' (CSB. trees) that are more cache conscious than B+trees. The goal of this paper is to make T-trees be cache conscious as CSB-trees. We propose a new index structure called a 'Cache Sensitive T-trees (CST-trees)'. We implemented CST-trees and compared performance of CST-trees with performance of other index structures.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.2
• /
• pp.37-46
• /
• 1996

This paper proposes a cost-effective solution to the synonym problem of virtual caches. In the proposed solution, a minimal hardware addition guarantees the correctness whereas the software counterpart helps improve the performance. The key to this proposed solution is an addition of a small physically-indexed cache called U-cache. The U-cache maintains the reverse translation information of the cache blocks that belong to unaligned virtual pages only, where aligned measns that the lower bits of the virtual page number match those of the corresponding physical page number. The page alignment is a simple software optimization to improve the performance of the U-cche hardware. With the combination of both hardware and software, the proposed solution reduces the hardware costs and minimizes software modification and performance degradation. Performance evaluation base on ATUM traces shows that a U-cache, with only a few entries, performs almost as well as fully-configured hardware-based solution when more than 95% of the pages are aligned.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.2
• /
• pp.78-82
• /
• 2015

General-purpose graphics processing units (GPGPUs) provide tremendous computational and processing power. Despite the latency hiding mechanism, a GPU architecture requires high memory bandwidth and lower latency between computational units and the memory system. For this reason, the current GPU architecture has private L1 caches in each core and a shared L2 cache to increase performance by reducing memory latency. But in some cases, this CPU-like cache design is not suitable for GPGPUs. In this paper, we analyze detailed cache performance related to GPGPU application characteristics, and suggest technical alternatives for the GPGPU architecture as future work.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.10
• /
• pp.1-9
• /
• 2018

Modern graphics processing units (GPUs) have become one of the most attractive platforms in exploiting high thread level parallelism with the support of new programming tools such as CUDA and OpenCL. Recent GPUs has applied cache hierarchy to support irregular memory access patterns; however, L1 data cache (L1D) exhibits poor efficiency in the GPU. This paper shows that the L1D does not always positively affect the applications in terms of performance and energy efficiency for the GPU. The performance of the GPU is even harmed by using the L1D for lots of applications. Our proposed technique exploits the characteristics of the currently-executed applications to predict the performance impact of the L1D on the GPU and then decides whether to continuously use the cache for the application or not. Our experimental results show that the proposed technique improves the GPU performance by 9.4% and saves up to 52.1% of the power consumption in the L1D.

• Journal of the Korea Society for Simulation
• /
• v.12 no.2
• /
• pp.35-44
• /
• 2003

The effectiveness of buffer cache replacement algorithms is critical to the performance of I/O systems. In this paper, we propose the degree of inter-reference gap (DIG) based block replacement scheme that retains merits of the least recently used (LRU) such as simple implementation and good cache hit ratio (CHR) for general patterns of references, and improves CHR further. In the proposed scheme, cache blocks with low DIGs are distinguished from blocks with high DIGs and the replacement block is selected among high DIGs blocks as done in the low inter-reference recency set (LIRS) scheme. Thus, by having the effect of the partitioning the cache memory dynamically based on DIGs, CHR is improved. Trace-driven simulation is employed to verified the superiority of the DIG based scheme and shows that the performance improves up to about 175% compared to the LRU scheme and 3% compared to the LIRS scheme for the same traces.