~/talk

Managing Memory Tiers with CXL in Virtualized Environments

links:none on file
links:none on file
links:none on file
name:Ryan Wee
links:none on file
links:none on file
links:none on file
links:none on file
links:none on file
links:none on file
links:none on file
slides:none on file

Introduces Intel Flat Memory Mode, a hardware-managed, cacheline-granular CXL tier that acts as a direct-map exclusive cache with zero CPU overhead and huge-page compatibility, evaluated on a preproduction Xeon (Granite Rapids) and an Astera Labs CXL card across 115 cloud workloads; hardware tiering gives near-DRAM performance for >82% of workloads but produces two failure modes — cross-VM contention and intra-VM conflicting access patterns — causing up to 34% slowdowns. Adds Memstrata, a hypervisor allocator using page coloring and an online slowdown estimator (L3 miss-latency proxy) to assign dedicated local DRAM, reducing outlier slowdowns to under ~6%.

AI-written summary · about

  1. 0:00Introduction and motivation
  2. 1:23CXL memory tiering challenges
  3. 2:59Limitations of software tiering
  4. 4:43Software tiering drawbacks
  5. 7:04Intel flat memory mode
  6. 9:30Addressing hardware limitations
  7. 11:31Memstrata and VM isolation
  8. 13:58Evaluation and conclusion
~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.