Computer+Architecture+and+Memory+Systems+Laboratory+%28CAMEL%29
-
LightPC Presents a Resilient System Using Only Non-Volatile Memory
Lightweight Persistence Centric System (LightPC) ensures both data and execution persistence for energy-efficient full system persistence
A KAIST research team has developed hardware and software technology that ensures both data and execution persistence. The Lightweight Persistence Centric System (LightPC) makes the systems resilient against power failures by utilizing only non-volatile memory as the main memory.
“We mounted non-volatile memory on a system board prototype and created an operating system to verify the effectiveness of LightPC,” said Professor Myoungsoo Jung. The team confirmed that LightPC validated its execution while powering up and down in the middle of execution, showing up to eight times more memory, 4.3 times faster application execution, and 73% lower power consumption compared to traditional systems.
Professor Jung said that LightPC can be utilized in a variety of fields such as data centers and high-performance computing to provide large-capacity memory, high performance, low power consumption, and service reliability.
In general, power failures on legacy systems can lead to the loss of data stored in the DRAM-based main memory. Unlike volatile memory such as DRAM, non-volatile memory can retain its data without power. Although non-volatile memory has the characteristics of lower power consumption and larger capacity than DRAM, non-volatile memory is typically used for the task of secondary storage due to its lower write performance. For this reason, nonvolatile memory is often used with DRAM. However, modern systems employing non-volatile memory-based main memory experience unexpected performance degradation due to the complicated memory microarchitecture.
To enable both data and execution persistent in legacy systems, it is necessary to transfer the data from the volatile memory to the non-volatile memory. Checkpointing is one possible solution. It periodically transfers the data in preparation for a sudden power failure. While this technology is essential for ensuring high mobility and reliability for users, checkpointing also has fatal drawbacks. It takes additional time and power to move data and requires a data recovery process as well as restarting the system.
In order to address these issues, the research team developed a processor and memory controller to raise the performance of non-volatile memory-only memory. LightPC matches the performance of DRAM by minimizing the internal volatile memory components from non-volatile memory, exposing the non-volatile memory (PRAM) media to the host, and increasing parallelism to service on-the-fly requests as soon as possible.
The team also presented operating system technology that quickly makes execution states of running processes persistent without the need for a checkpointing process. The operating system prevents all modifications to execution states and data by keeping all program executions idle before transferring data in order to support consistency within a period much shorter than the standard power hold-up time of about 16 minutes. For consistency, when the power is recovered, the computer almost immediately revives itself and re-executes all the offline processes immediately without the need for a boot process.
The researchers will present their work (LightPC: Hardware and Software Co-Design for Energy-Efficient Full System Persistence) at the International Symposium on Computer Architecture (ISCA) 2022 in New York in June. More information is available at the CAMELab website (http://camelab.org).
-Profile: Professor Myoungsoo Jung Computer Architecture and Memory Systems Laboratory (CAMEL)http://camelab.org School of Electrical EngineeringKAIST
2022.04.25 View 18508 -
CXL-Based Memory Disaggregation Technology Opens Up a New Direction for Big Data Solution Frameworks
A KAIST team’s compute express link (CXL) provides new insights on memory disaggregation and ensures direct access and high-performance capabilities
A team from the Computer Architecture and Memory Systems Laboratory (CAMEL) at KAIST presented a new compute express link (CXL) solution whose directly accessible, and high-performance memory disaggregation opens new directions for big data memory processing. Professor Myoungsoo Jung said the team’s technology significantly improves performance compared to existing remote direct memory access (RDMA)-based memory disaggregation.
CXL is a peripheral component interconnect-express (PCIe)-based new dynamic multi-protocol made for efficiently utilizing memory devices and accelerators. Many enterprise data centers and memory vendors are paying attention to it as the next-generation multi-protocol for the era of big data.
Emerging big data applications such as machine learning, graph analytics, and in-memory databases require large memory capacities. However, scaling out the memory capacity via a prior memory interface like double data rate (DDR) is limited by the number of the central processing units (CPUs) and memory controllers. Therefore, memory disaggregation, which allows connecting a host to another host’s memory or memory nodes, has appeared.
RDMA is a way that a host can directly access another host’s memory via InfiniBand, the commonly used network protocol in data centers. Nowadays, most existing memory disaggregation technologies employ RDMA to get a large memory capacity. As a result, a host can share another host’s memory by transferring the data between local and remote memory.
Although RDMA-based memory disaggregation provides a large memory capacity to a host, two critical problems exist. First, scaling out the memory still needs an extra CPU to be added. Since passive memory such as dynamic random-access memory (DRAM), cannot operate by itself, it should be controlled by the CPU. Second, redundant data copies and software fabric interventions for RDMA-based memory disaggregation cause longer access latency. For example, remote memory access latency in RDMA-based memory disaggregation is multiple orders of magnitude longer than local memory access.
To address these issues, Professor Jung’s team developed the CXL-based memory disaggregation framework, including CXL-enabled customized CPUs, CXL devices, CXL switches, and CXL-aware operating system modules. The team’s CXL device is a pure passive and directly accessible memory node that contains multiple DRAM dual inline memory modules (DIMMs) and a CXL memory controller. Since the CXL memory controller supports the memory in the CXL device, a host can utilize the memory node without processor or software intervention. The team’s CXL switch enables scaling out a host’s memory capacity by hierarchically connecting multiple CXL devices to the CXL switch allowing more than hundreds of devices. Atop the switches and devices, the team’s CXL-enabled operating system removes redundant data copy and protocol conversion exhibited by conventional RDMA, which can significantly decrease access latency to the memory nodes.
In a test comparing loading 64B (cacheline) data from memory pooling devices, CXL-based memory disaggregation showed 8.2 times higher data load performance than RDMA-based memory disaggregation and even similar performance to local DRAM memory. In the team’s evaluations for a big data benchmark such as a machine learning-based test, CXL-based memory disaggregation technology also showed a maximum of 3.7 times higher performance than prior RDMA-based memory disaggregation technologies.
“Escaping from the conventional RDMA-based memory disaggregation, our CXL-based memory disaggregation framework can provide high scalability and performance for diverse datacenters and cloud service infrastructures,” said Professor Jung. He went on to stress, “Our CXL-based memory disaggregation research will bring about a new paradigm for memory solutions that will lead the era of big data.”
-Profile: Professor Myoungsoo Jung Computer Architecture and Memory Systems Laboratory (CAMEL)http://camelab.org School of Electrical EngineeringKAIST
2022.03.16 View 16612