Recent Projects
I/O Performance Improvement in HPC Manycore Environment
As modern HPC applications generate massive amounts of data, there is an increasing demand for understanding supercomputer system with manycore processor such as Intel Knights Landing. I/O performance analysis and improvement should be made and finding suitable configurations for various environments with KNL processor is needed. In addition, as the cost of high performance storage[…]
Blockchain
Blockchain is designed to be consistent and stable in an unreliable and distributed environment, allowing end users to perform contracts without intermediate entities through smart devices. However, there are several performance issues with the blockchain. We improve the performance of blockchain, privacy problems and wallet management, based on Ethereum Blockchain, a representative system of blockchain,[…]
FlexGPU
The graphics processing unit (GPU) is extensively used in diverse domains, such as finance, machine learning, and image processing. The GPU can be underutilized as multiple applications may not share the same GPU concurrently owing to a memory oversubscription issue. For example, when applications that require fewer computational resources but a larger GPU memory are[…]
HPC Resource Utilization
High performance computing (HPC) is becoming a promising technology due to its high computing power and efficiency. As the amount of data needs to be processed increases, Big Data and Machine learning applications are utilizing HPC system for fast execution time. In HPC system, distributed file system such as lustre and ceph are being used[…]
Bridging the Semantic Gap Between OS and Middleware
Bridging the semantic gap between operating system and middleware General-purpose operating systems (OSs) are designed to provide system resources to applications running on a system in a fair manner at system-level. However, a single application may find it difficult to fully exploit the system’s best performance due to this system-level fairness. Middleware, such as database[…]
Rack Scale Architecture Computing for Server Resource Utilization
The growth rate of cloud data centers is growing and the importance of data centers is increasing. However, it can be known that the servers that provide these cloud services have very low server utilization. The current data centers of the second generation clouds have limit to allocate resources to workload dynamically. It leads to[…]
Zero-Copy Memory Move by Page Mapping Change
As time passes, data which is processed by computers is getting bigger and bigger. And capacity and bandwidth of memory also are getting bigger. Improving performance of devices to process more data is a simple and fundamental way. But there is a way by software improvement to increase data processing performance too. And we try[…]
Distributed File System
As big data and cloud computing are thriving in IT ecosystem, it is becoming more and more important for enterprises to collect and analyze huge amount of user data. To process big data, it is fundamentally required to efficiently store and manage them, which necessitate distributed file system for this purpose. GlusterFS and[…]
Cache Partitioning
In multi-core system where different cores share last-level cache, performance interference between processes become a real issue. Our lab is working on researches to reduce this performance interference. Our lab is working on researches to reduce this performance interference between processes, which can lower the overall performance of the system as well as making it[…]
Performance Isolation for Datacenters
Datacenters are becoming more important as they are heavily used web or big data workloads in companies such as Facebook, Google, Amazon, etc. For efficient utlization of datacenters, it is required to solve performance problem due to interference between service workloads. Performance interference is inevitable in datacenters where many services compete for limited resources; to[…]
Multi-core Scalability
As we entered the 21st century, improving processor performance through faster clock speed has reached the wall due to power and heat problem. Therefore, modern processors started to adopt multi-core systems, which house multiple processor cores in one system. It was relatively easy to scale a program on a different processor that has higher clock speed[…]
SSD Optimization
Storage device is a fundamental element of computer systems that holds user data. The advance of storage hardware and big data platforms also elicited active researches on storage stack. Our lab is doing research on improving the performance and stability of storage devices by optimizing software technology, which is critical in enhancing user experience of computing[…]