While virtualization has become an important component in VDI, I/O is still an open challenge in the virtualization. Although several researches attempted to mitigate virtualized I/O overheads, many of them concentrated on either VM-level optimization or hypervisor-level one, not both. In this paper, we propose a multi-layered cache implementation, called MultiCache, which combines the guest-level optimization with the hypervisor-level I/O optimization. We executed the performance measurement of MultiCache to verify its effectiveness.
Taehoon KimJaechun NoZhegao PiaoSeong Joon Yoo
Taehoon KimKwanghyeok SongJaechun NoSungSoon Park
Taehoon KimSeungho ChoiJaechun NoSung‐Soon Park
Jin-Suk MaHak-Young KimWan Choi
Ming-ting WeiYu‐Shiang LinChe–Rung Lee