• ETRI Journal
• /
• 제39권3호
• /
• pp.428-436
• /
• 2017

State-of-the-art processors require increasingly complicated memory services for high performance and low power consumption. In particular, they request transfers within a burst in a wrap-around order to minimize the miss penalty of a cache. However, synchronous dynamic random access memories (SDRAMs) do not always generate transfers in the wrap-round order required by the processors. Thus, a memory subsystem rearranges the SDRAM transfers in the wrap-around order, but the rearrangement process may increase memory latency and waste the bandwidth of on-chip interconnects. In this paper, we present a memory subsystem that is effective for the wrapping bursts of a cache. The proposed memory subsystem makes SDRAMs generate transfers in an intermediate order, where the transfers are rearranged in the wrap-around order with minimal penalties. Then, the transfers are delivered with priority, depending on the program locality in space. Experimental results showed that the proposed memory subsystem minimizes the memory performance loss resulting from wrapping bursts and, thus, improves program execution time.

• International journal of advanced smart convergence
• /
• 제12권4호
• /
• pp.164-170
• /
• 2023

With the recent advances in AI (artificial intelligence) and HPC (high-performance computing) technologies, deep learning is proliferated in various domains of the 4th industrial revolution. As the workload volume of deep learning increasingly grows, analyzing the memory reference characteristics becomes important. In this article, we analyze the memory reference traces of deep learning workloads in comparison with traditional workloads specially focusing on read and write operations. Based on our analysis, we observe some unique characteristics of deep learning memory references that are quite different from traditional workloads. First, when comparing instruction and data references, instruction reference accounts for a little portion in deep learning workloads. Second, when comparing read and write, write reference accounts for a majority of memory references, which is also different from traditional workloads. Third, although write references are dominant, it exhibits low reference skewness compared to traditional workloads. Specifically, the skew factor of write references is small compared to traditional workloads. We expect that the analysis performed in this article will be helpful in efficiently designing memory management systems for deep learning workloads.

• ETRI Journal
• /
• 제33권6호
• /
• pp.904-913
• /
• 2011

Three-dimensional (3D) memories using through-silicon vias (TSVs) as vertical buses across memory layers will likely be the first commercial application of 3D integrated circuit technology. The memory dies to stack together in a 3D memory are selected by a die-selection method. The conventional die-selection methods do not result in a high-enough yields of 3D memories because 3D memories are typically composed of known-good-dies (KGDs), which are repaired using self-contained redundancies. In 3D memory, redundancy sharing between neighboring vertical memory dies using TSVs is an effective strategy for yield enhancement. With the redundancy sharing strategy, a known-bad-die (KBD) possibly becomes a KGD after bonding. In this paper, we propose a novel die-selection method using KBDs as well as KGDs for yield enhancement in 3D memory. The proposed die-selection method uses three search-space conditions, which can reduce the search space for selecting memory dies to manufacture 3D memories. Simulation results show that the proposed die-selection method can significantly improve the yield of 3D memories in various fault distributions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권11호
• /
• pp.5234-5251
• /
• 2018

It is well-known that virtualization technology can bring many benefits not only to users but also to service providers. From the view of system security and resource utility, higher resource sharing degree and higher system reliability can be obtained by the introduction of virtualization technology in distributed cloud. The small size time-sharing multiplexing technology which is based on virtual machine in distributed cloud platform can enhance the resource utilization effectively by server consolidation. In this paper, the concept of memory block and user satisfaction is redefined combined with user requirements. According to the unbalanced memory resource states and user preference requirements in multi-virtual machine environments, a model of proper memory resource allocation is proposed combined with memory block and user satisfaction, and at the same time a memory optimization allocation algorithm is proposed which is based on virtual memory block, makespan and user satisfaction under the premise of an orderly physical nodes states also. In the algorithm, a memory optimal problem can be transformed into a resource workload balance problem. All the virtual machine tasks are simulated in Cloudsim platform. And the experimental results show that the problem of virtual machine memory resource allocation can be solved flexibly and efficiently.

• International Journal of Internet, Broadcasting and Communication
• /
• 제9권1호
• /
• pp.29-34
• /
• 2017

Over the past several decades, embedded system and flight control computer technologies have been evolved to meet the diverse needs of the mobile device market. Current embedded systems are at the heart of technologies that can take advantage of small-sized specialized hardware while still providing high-efficiency performance at low cost. One of these key technologies is multiple memory banks. For example, a dual memory bank can provide two times more memory bandwidth in the same memory space. This benefit take lower cost to provide the same bandwidth. However, there is still few software technologies to support the efficient use of multiple memory banks. In this study, we present a technique to efficiently exploit multiple memory banks by software support. Specifically, our technique use an interference graph to optimally allocate data to different memory banks by an optimizing compiler. As a result, the execution time can be improved upto 7% with the proposed technique.

• 대한임베디드공학회논문지
• /
• 제11권2호
• /
• pp.57-65
• /
• 2016

In this paper, we propose a system for efficient use of shared memory between CPU and GPU. The system, called Fusion Architecture, assures consistency of the shared memory and minimizes cache misses that frequently occurs on Heterogeneous System Architecture or Unified Virtual Memory based systems. It also maximizes the performance for memory intensive jobs by efficient allocation of GPU cores. To test between architectures on various scenarios, we introduce the Fusion Architecture Analyzer, which compares OpenMP, OpenCL, CUDA, and the proposed architecture in terms of memory overhead and process time. As a result, Proposed fusion architectures show that the Fusion Architecture runs benchmarks 55% faster and reduces memory overheads by 220% in average.

• International Journal of Internet, Broadcasting and Communication
• /
• 제14권1호
• /
• pp.188-193
• /
• 2022

The energy efficiency of memory systems is an important task in designing future computer systems as memory capacity continues to increase to accommodate the growing big data. In this article, we present an energy-efficient last-level cache management policy for future mobile systems. The proposed policy makes use of low-power PCM (phase-change memory) as the main memory medium, and reduces the amount of data written to PCM, thereby saving memory energy consumptions. To do so, the policy keeps track of the modified cache lines within each cache block, and replaces the last-level cache block that incurs the smallest PCM writing upon cache replacement requests. Also, the policy considers the access bit of cache blocks along with the cache line modifications in order not to degrade the cache hit ratio. Simulation experiments using SPEC benchmarks show that the proposed policy reduces the power consumption of PCM memory by 22.7% on average without degrading performances.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.258-262
• /
• 2009

In this paper, we propose an effective memory reduction algorithm to reduce the amount of reference frame buffer and memory bandwidth in video encoder and decoder. In general video codecs, decoded previous frames should be stored and referred to reduce temporal redundancy. Recently, reference frames are recompressed for memory efficiency and bandwidth reduction between a main processor and external memory. However, these algorithms could hurt coding efficiency. Several algorithms have been proposed to reduce the amount of reference memory with minimum quality degradation. They still suffer from quality degradation with fixed-bit allocation. In this paper, we propose an adaptive block-based min-max quantization that considers local characteristics of image. In the proposed algorithm, basic process unit is $8{\times}8$ for memory alignment and apply an adaptive quantization to each $4{\times}4$ block for minimizing quality degradation. We found that the proposed algorithm could improve approximately 37.5% in coding efficiency, compared with an existing memory reduction algorithm, at the same memory reduction rate.

• 대한간호학회지
• /
• 제47권3호
• /
• pp.431-444
• /
• 2017

Purpose: This study was conducted to verify the effects of a memory and visual-motor integration program for older adults based on self-efficacy theory. Methods: A non-equivalent control group pretest-posttest design was implemented in this quasi-experimental study. The participants were 62 older adults from senior centers and older adult welfare facilities in D and G city (Experimental group=30, Control group=32). The experimental group took part in a 12-session memory and visual-motor integration program over 6 weeks. Data regarding memory self-efficacy, memory, visual-motor integration, and depression were collected from July to October of 2014 and analyzed with independent t-test and Mann-Whitney U test using PASW Statistics (SPSS) 18.0 to determine the effects of the interventions. Results: Memory self-efficacy (t=2.20, p=.031), memory (Z=-2.92, p=.004), and visual-motor integration (Z=-2.49, p=.013) increased significantly in the experimental group as compared to the control group. However, depression (Z=-0.90, p=.367) did not decrease significantly. Conclusion: This program is effective for increasing memory, visual-motor integration, and memory self-efficacy in older adults. Therefore, it can be used to improve cognition and prevent dementia in older adults.

• 한국콘텐츠학회논문지
• /
• 제9권2호
• /
• pp.163-172
• /
• 2009

문화콘텐츠가 문화산업의 중심으로 부상하면서, 콘텐츠와 이를 활용한 문화산업에 대한 관심이 어느 때 보다 높다. 문화는 기억을 중심으로 이루어지며, 기억을 배제한 문화란 존재할 수 없다. 기억은 개인의 차원을 넘어 집단적, 사회적 기억으로 작용한다. 또한 문화는 기억을 담지할 매체를 필연적으로 요구하는데, 최근의 영상텍스트는 새로운 기억매체로서 주목받고 있다. 이러한 점에서 이 논문은 사회 문화적 기억으로서 '기억'의 의미를 탐색해 보고, 중국 6세대 영화를 중심으로 사회 문화적 기억과 그것을 담지해내는 영상텍스트의 기억복원의 의미를 분석하는 데 초점을 두었다. 또한 '대항기억'으로서 '개인기억'의 문화적 의의를 탐색하고, 이를 통해 기억산업과 콘텐츠의 접합 가능성을 찾아보는 시도를 모색했다. 6세대 영화를 주목한 이유는 중국 당국이 제시하는 '공식기억'에 대항하며 '지하'에서 만들어졌던 이들 영화가 국제적인 영화제에서 괄목한 만한 성과를 획득해 나가고 있기 때문이다.