속성:Applicability
외관
이 메모가 맞는 조건과 예외를 적습니다.
a
archer adaptive memory compression with page association rule awareness for high speed response 98bcc760 +
Mobile anonymous memory with recurring or correlated page content.
Limits: Correlation discovery and benefits depend on app mix; results target mobile workloads. +
zram과 flash page cache를 쓰는 모바일 OS의 hot app switching에 높음. +
서로 다른 SATA/NVMe SSD를 포함한 parity RAID 설계 탐색에 적용합니다.
Limits: HotStorage 2024 제안·동기 연구이며 Asym-RAID 구현, fault tolerance, rebuild, dynamic heterogeneity의 end-to-end 평가는 future work입니다. +
Page cache를 유지해야 하는 초저지연 NVMe용 커널 I/O 경로에 직접 적용 가능성이 높다. +
GPU applications with direct SSD access and overlapable computation.
Limits: Requires GPU/SSD integration and enough independent work; proactive access is easier for predictable patterns. +
OS·network stack·crypto 코드 분석.
Limits: 분석 가능한 코드·입력 모델·instrumentation coverage와 projector 정확도에 의존한다. +
Flash-backed databases and buffer caches under dirty-page pressure.
Limits: Requires extra DRAM and careful ordering/recovery; value falls when dirty-victim stalls are rare. +
aware automate workload autoscaling with reinforcement learning in production cloud systems cb18da5e +
클라우드 서비스 오토스케일링.
Limits: 보상 설계·시뮬레이터 전이·운영 안전성 검증에 의존한다. +
b
Rowhammer·DDR PRAC 연구 맵핑에는 관련되지만 기술 의사결정에는 아직 사용할 수 없다. +
vLLM, ZeroMQ, Unix-domain socket runtimes, and nested workspaces where general temp storage and IPC path constraints differ. +
복합적인 On-CPU와 대기 병목을 가진 서버·시스템 응용의 성능 분석. +
DRAM systems deploying preventive RowHammer mitigations.
Limits: Relies on correct culprit identification and simulated mitigation models; benign high-activity threads can be throttled. +
Shared NVMe SSDs serving unmodified storage applications.
Limits: Depends on suitable IOMMU/device support and a setup/control path in the OS. +
byteexpress a high performance and traffic efficient inline transfer of small payloads over nvme a7c01a49 +
NVMe workloads dominated by small reads/writes on modifiable controllers.
Limits: Prototype uses OpenSSD and targets sub-page payloads; benefit does not generalize to large transfers. +
c
Multimedia buffer를 application과 device가 반복 공유하는 mobile SoC에 해당한다. +
RAG services with recurring document chunks and reusable KV caches.
Limits: Requires cache storage and token selection; benefits depend on chunk reuse and retrieval composition. +
EBS 위 LSM KV/DB.
Limits: EBS와 선정 워크로드 중심이며 캐시·IOPS 정책의 이식성은 미검증이다. +
CacheLib-like persistent flash caches on ZNS SSDs.
Limits: Workshop exploration with three schemes; results depend on cache-unit/zone sizes, workload locality, ZNS device, and recovery design. +
mixed real-time/best-effort embedded Linux workload에 해당한다. +
On-device LLM inference on unified-memory mobile SoCs.
Limits: Results are tied to one modern Snapdragon platform and its GPU/NPU software stack. +