속성으로 검색

이 문서는 속성과 이름이 지정된 값으로 설명된 개체를 찾기 위한 간단한 탐색 인터페이스를 제공합니다. 그 밖에 사용 가능한 검색 인터페이스에는 문서 속성 검색 및 Ask 쿼리 빌더가 있습니다.

결과 목록

• Lesson:pivot-b200-20260717-portable-pipeline-hash-a01  + (Content-addressed experiment code should hash normalized repository- or package-relative names plus bytes. Test identical trees at different absolute roots and fail closed on filename/content changes or an empty code set.)
• Lesson:tiered memory management access latency is the key aaf24437  + (Control placement with measured queueing-sensitive latency and balance load across tiers.)
• Lesson:swam revisiting swap and oomk for improving application responsiveness on mobile devices 79b3b8a3  + (Coordinate graceful reclamation and last-resort eviction using the future restart cost of each workload.)
• Lesson:ice collaborating memory and process management for user experience on resource limited mobile d 312e453f  + (Coordinate reclamation with execution control when background activity recreates the state being reclaimed.)
• Lesson:efficient memory overcommitment for i o passthrough enabled vms via fine grained page meta data d32f4845  + (DMA 환경의 메모리 회수는 페이지별 소유·사용 상태 추적이 핵심이다.)
• Lesson:technical review dram 0a5cae63  + (DRAM-less FTL에서는 전역 유효 페이지 메타데이터를 별도로 저장하기보다 이미 존재하는 L2P 세그먼트와 함께 압축·지역화하면 조회 I/O를 줄일 수 있다.)
• Lesson:dram 8d56adf1  + (DRAM-less FTL에서는 전역 유효 페이지 메타데이터를 별도로 저장하기보다 이미 존재하는 L2P 세그먼트와 함께 압축·지역화하면 조회 I/O를 줄일 수 있다.)
• Lesson:technical review cache scheme of shared buffer mappings for energy efficiency of mobile devices f1dee859  + (Data 자체뿐 아니라 비싼 address-translation 또는 mapping 결과도 identity와 lifetime이 안전하다면 재사용할 수 있다.)
• Lesson:cache scheme of shared buffer mappings for energy efficiency of mobile devices 4b6db791  + (Data 자체뿐 아니라 비싼 address-translation 또는 mapping 결과도 identity와 lifetime이 안전하다면 재사용할 수 있다.)
• Lesson:scalable far memory balancing faults and evictions edcf1536  + (Decouple and pipeline opposing memory flows, prioritizing scalable progress over perfect victim selection.)
• Lesson:stem streaming based fpga acceleration for large scale compactions in lsm kv d204166c  + (Design data-reorganization accelerators around streaming partitions and dynamically scheduled units.)
• Lesson:dlora dynamically orchestrating requests and adapters for lora llm serving 3827bfa3  + (Dynamically switch between general and specialized execution paths using observed demand and switching cost.)
• Lesson:ozz identifying kernel out of order concurrency bugs with in vivo memory access reordering 0b85f639  + (Emulate nondeterministic low-level behavior and explore it jointly with software scheduling.)

• Lesson:citadel rethinking memory allocation to safeguard against inter domain rowhammer exploits 8fd39aca  + (Encode hardware fault geometry directly into allocation domains.)

• Lesson:pact a criticality first design for tiered memory 20db3445  + (Estimate per-item criticality from exposed latency and parallelism, then design migration policies around its skew.)
• Lesson:understanding and mitigating covert and side channel vulnerabilities introduced by rowhammer def 0e272653  + (Evaluate security mechanisms for timing/noninterference, not only for prevention efficacy.)
• Lesson:crossprefetch accelerating i o prefetching for modern storage 59296dd1  + (Expose application intent to lower layers and prevent speculative work from delaying demand work.)
• Lesson:asynchrony and gpus bridging this dichotomy for i o with agio 6fab41bf  + (Expose nonblocking initiation plus device-local progress and continuation mechanisms.)
• Lesson:harvesting memory bound cpu stall cycles in software with msh 5866da08  + (Expose resource-stall opportunities to a policy-controlled runtime when hardware sharing is too coarse.)
• Lesson:constructing and analyzing the lsm compaction design space 1aa4ace5  + (Factor a policy into orthogonal primitives before comparing implementations.)
• Lesson:research autopilot 20260720t030001z gpu  + (For cache-capacity models, measure both nested tensor boundaries and score held-out tails; a sharp output-tensor transition can coexist with a graded working-set rise, while an unmodified square-only harness cannot answer rectangular reload-law claims.)
• Lesson:cold archives must materialize content addressed model files and cross a storage failure domain 03687d8e  + (For cold reconstruction, archive the exact … For cold reconstruction, archive the exact source commit, raw decisions/journals, frozen datasets, locator manifests, and only the locator-verified model files materialized as regular files. Seal a sorted content manifest, verify it after publication, and do not declare preservation complete until the same manifest passes in a storage-failure domain that will outlive the compute host.domain that will outlive the compute host.)
• Lesson:pivot-b200-20260717-immutable-runtime-a01  + (For editable or precompiled ML runtimes, s … For editable or precompiled ML runtimes, snapshot the entire effective package rather than only Git files or shared objects; reject inherited behavior controls, prove actual import origin, content-bind both package and native subsets, and retain a strict sanitized audit with the result. a strict sanitized audit with the result.)
• Lesson:pivot-b200-20260717-native-extension-boundary-a01  + (For environment-linked native modules, has … For environment-linked native modules, hash logical in-repository module identity plus followed target bytes, exclude absolute target paths from the digest, fail closed on broken or non-regular targets, and recompute the digest at suite completion to close the normal time-of-check/time-of-use gap. the normal time-of-check/time-of-use gap.)
• Lesson:loom efficient capture and querying of high frequency telemetry bbcbd611  + (For high-rate observability, bound write-path index work, index immutable chunks sparsely, and let queries scan a small active/unindexed tail instead of sampling away rare events.)
• Lesson:fleetio managing multi tenant cloud storage with multi agent reinforcement learning 714ca9fb  + (For multi-objective resource sharing, condition policies and rewards on workload class and expose actions directly to the scheduler.)
• Lesson:exploring the asynchrony of slow memory filesystem with easyio 0952deee  + (For slow-memory filesystems, co-design asynchronous DMA, task scheduling, update ordering, locking, and channel QoS; do not transpose the result to GPU copy engines without direct evidence.)
• Lesson:klotski efficient mixture of expert inference via expert aware multi batch pipeline bae4ea7b  + (For sparse models, schedule batches by the union and sequence of resources they activate, not batch size alone.)
• Lesson:pivot-b200-20260717-tmux-argument-boundary-a01  + (For tmux send-keys through SSH, do not ass … For tmux send-keys through SSH, do not assume local one-argument quoting survives. Encode separators explicitly as tmux Space key tokens or preserve quoting inside the remote command string, then immediately verify the pane, lock owner, run directory, and GPU state. lock owner, run directory, and GPU state.)
• Lesson:pod attention unlocking full prefill decode overlap for faster llm inference caae5b3e  + (Fuse and partition complementary kernels at the SM level rather than only batching at request level.)
• Lesson:technical review performance optimization of object tracking algorithms in opencv on gpus 09f20f75  + (GPU 최적화는 연산식만 바꾸기보다 register pressure→occupancy와 다단계 kernel launch/transfer를 함께 측정해 메모리 계층 이동과 kernel fusion을 선택해야 한다.)
• Lesson:performance optimization of object tracking algorithms in opencv on gpus 8158d2ec  + (GPU 최적화는 연산식만 바꾸기보다 register pressure→occupancy와 다단계 kernel launch/transfer를 함께 측정해 메모리 계층 이동과 kernel fusion을 선택해야 한다.)
• Lesson:asymmetric raid rethinking raid for ssd heterogeneity bfd97257  + (Heterogeneous RAID 설계에서는 장치별 capacity와 performance를 stripe-group/LBA 배치에 명시적으로 반영하되, prototype 평가 전에는 동기 microbenchmark를 제안 시스템의 이득으로 해석하지 않는다.)
• Lesson:technical review exploiting asymmetric cpu performance for fast startup of subsystem in mobile s f33a336a  + (Heterogeneous SoC에서는 steady-state execution 위치와 initialization 실행 위치를 분리할 수 있다.)
• Lesson:exploiting asymmetric cpu performance for fast startup of subsystem in mobile smart devices c00616ec  + (Heterogeneous SoC에서는 steady-state execution 위치와 initialization 실행 위치를 분리할 수 있다.)
• Lesson:tiered memory management beyond hotness 6ad152df  + (Incorporate latency and parallelism into memory criticality, then separate initial placement from runtime correction.)
• Lesson:byteexpress a high performance and traffic efficient inline transfer of small payloads over nvme a7c01a49  + (Inline tiny values into control messages when protocol headroom and ordering permit.)
• Lesson:mitigating application resource overload with targeted task cancellation 90270bf8  + (Instrument application-defined resources and existing cancellation initiators; choose victims by released contended load, and track throughput, tail latency, drop rate, and cancellation safety together.)
• Lesson:emt an os framework for new memory translation architectures 6d3356f6  + (Introduce a stable translation interface before adding new hardware-specific structures.)