著者

Balazs Gerofi Hajime Fujita Yutaka Ishikawa

雑誌

情報処理学会論文誌コンピューティングシステム(ACS) (ISSN:18827829)

巻号頁・発行日

vol.3, no.1, pp.1-12, 2010-03-16

---

Single IP Address cluster offers a transparent view of a cluster of machines as if they were a single computer on the network. In such an environment, process migration can play a significant role for providing services seamlessly and for increasing sustainability. In this paper we propose a live migration mechanism which is capable of moving processes that maintain a massive amount of network connections, supporting both TCP and UDP sockets. Incoming packet loss during socket migration is prevented by exploiting the broadcast property of the Single IP Address cluster, while process live migration minimizes the execution freeze time during the actual migration of the process context. Performance evaluation on machines equipped with a 2.4 GHz CPU and Gigabit Ethernet interconnect shows that migrating a process of 1GB image size and over 1000 established network connections results in less than 200 ms process freeze time, rendering the transition fully transparent and responsive from the clients' point of view. The implementation is comprised entirely of a kernel module for Linux 2.6, without any changes to the existing kernel code.

著者

平井 浩一 小田和 友仁 岡本 高幸 二宮 温 住元 真司 高木 将通 Balazs Gerofi 山口 訓央 小倉 崇浩 亀山 豊久 堀 敦史 石川 裕

雑誌

研究報告システム・アーキテクチャ(ARC) (ISSN:21888574)

巻号頁・発行日

vol.2015-ARC-215, no.2, pp.1-8, 2015-05-19

---

将来の HPC 向けの OS としては,メニーコアへの最適化が必須となってきており,それを実現するための OS として McKernel を選択し,計算センターにおけるバッチジョブ運用への適応を進めている.本論文では,将来のスーパーコンピュータ上で,McKernel に適応したバッチジョブ運用を実現する場合の課題を述べ,現状の検討状況について述べる.

著者

Masamichi Takagi Balazs Gerofi Norio Yamaguchi Takahiro Ogura Toyohisa Kameyama Atsushi Hori Yutaka Ishikawa

雑誌

研究報告システム・アーキテクチャ(ARC) (ISSN:21888574)

巻号頁・発行日

vol.2015-ARC-215, no.1, pp.1-8, 2015-05-19

---

Processor core count in high-end computing has seen a steady increase during the past decade and next generation supercomputers will likely deploy many-core based systems. At the same time, from a software environment point of view, Linux-compatibility has become wide-spread in the High Performance Computing (HPC) domain. We consider the challenges of operating system (OS) design targeting next generation high-end computing. We believe that the most urging issues to be addressed are as follows. (1) Exploiting deep memory hierarchies, (2) Reducing cache pollution by OS services and minimizing OS noise, (3) Making it easy to design and deploy application specific kernels, (4) Providing a Linux compatible programming / run-time environment and (5) Enabling seamless tracking of upstream Linux kernel changes. We contend that existing approaches to HPC operating systems, which either employ a stripped down Linux environment or a specific light-weight kernel built from scratch, are not feasible to deal with these challenges. In this paper, we discuss the design decisions of our proposed hybrid kernel design for providing a Linux-compatible light-weight kernel.

著者

佐伯 裕治 清水 正明 白沢 智輝 中村 豪 高木 将通 Balazs Gerofi 思 敏 石川 裕 堀 敦史

出版者

一般社団法人情報処理学会

雑誌

研究報告システムソフトウェアとオペレーティング・システム(OS)

巻号頁・発行日

vol.2013, no.15, pp.1-7, 2013-04-18

---

メニーコアプロセッサ向けの OS として,Linux カーネルと軽量カーネルが連携して管理するヘテロジニアス構成の OS を開発している.軽量カーネル上においても Linux カーネルのシステムコールを提供するために,軽量カーネルで実現されない Linux システムコールの処理は Linux カーネルに委譲する.引数がデータ領域を示すシステムコールの場合,転送が必要なデータの構造は API 仕様に依存するため,300 種類以上の Linux 互換システムコールに個別に対応したデータ転送を実装する必要がある.本稿では,システムコール処理対象となるデータを同一仮想アドレスへのメモリマップを行う方式により,軽量カーネルに個々のシステムコール処理を実装することなく Linux カーネルに委譲する機構と,その基本評価結果について報告する.We have been developing a heterogeneous OS composed of Linux and lightweight kernels for manycore processor. In order to provide all Linux system calls in the lightweight kernel, those primitives which are not provided by the lightweight kernel are delegated to the Linux kernel. Each system call differs in the number of arguments and argument types, and thus the code transferring arguments and results is implemented in each delegating system call. It is impractical to implement all Linux APIs, i.e., more than 300 system calls. Therefore, we developed a delegation mechanism of system calls without individual implementation to pass the data between the lightweight kernel and Linux using a memory mapping technique. In this technique, a user-level virtual address space in the lightweight kernel is mapped to the same position in a Linux process. We report the result of basic evaluation of system calls on lightweight kernel developed on Intel(R) Xeon PhiTM Coprocessor.