sirjee
Re:Linux kernel cgroups writeback high CPU troubleshooting
on Friday February 14, 2025 @12:30PM (#43050722)
Usually Google applications uses high level network storage services like collosus, BigTable or spanner and these high level services are backed by dedicated storage appliances where they bypass the kernel for SSDs and use direct block IO for slow disks. For network, they are moving towards userspace network [1].
[1] https://research.google/pubs/snap-a-microkernel-approach-to-...
Re:Linux kernel cgroups writeback high CPU troubleshooting
on Friday February 14, 2025 @11:59AM (#43050427)
BTW company-1 == Google and company-2 == FB/Meta.
In addition, Google has completely removed local storage from their servers, so there is no disk I/O at all.
Re:Windows Memory Mapped File IO
on Saturday November 16, 2024 @01:31AM (#42154827)
Linux have madvise(MADV_DONTNEED).
Re:Transparent memory offloading: more memory at a fraction of the cos...
on Tuesday June 21, 2022 @02:10AM (#31820483)
How does this compare with Google's software defined far memory: https://research.google/pubs/pub48551/?
Re:Fuchsia Workstation
on Monday March 28, 2022 @02:56AM (#30827855)
Not sure I understand the relationship between real-time guarantees and performance.
Re:Google removes its head of diversity after 2007 blog post surfaces
on Thursday June 03, 2021 @03:50PM (#27386415)
He was conflating jews with Israeli government and zionists, so I can assume he was unread on the topic.
Re:Google removes its head of diversity after 2007 blog post surfaces
on Thursday June 03, 2021 @02:37PM (#27385506)
No we should not accuse but as the citizen of such a regime I should feel responsible of its actions, I should condemn it and may be expect accusations of supporting such a regime.
Re:Google removes its head of diversity after 2007 blog post surfaces
on Thursday June 03, 2021 @01:30PM (#27384598)
Pretty sure he was not chief diversity officer 14 years ago but I agree it should have been "If I were an israeli...".