Optics in computing: From photonic network-on-chip to chip-to-chip interconnects and disintegrated architectures
Following a decade of radical advances in the areas of integrated photonics and computing
architectures, we discuss the use of optics in the current computing landscape attempting to …
architectures, we discuss the use of optics in the current computing landscape attempting to …
Designing far memory data structures: Think outside the box
Technologies like RDMA and Gen-Z, which give access to memory outside the box, are
gaining in popularity. These technologies provide the abstraction of far memory, where …
gaining in popularity. These technologies provide the abstraction of far memory, where …
Spandex: A flexible interface for efficient heterogeneous coherence
Recent heterogeneous architectures have trended toward tighter integration and shared
memory largely due to the efficient communication and programmability enabled by this …
memory largely due to the efficient communication and programmability enabled by this …
Exploiting commutativity to reduce the cost of updates to shared data in cache-coherent systems
We present Coup, a technique to lower the cost of updates to shared data in cache-coherent
systems. Coup exploits the insight that many update operations, such as additions and …
systems. Coup exploits the insight that many update operations, such as additions and …
Revisiting the complexity of hardware cache coherence and some implications
Cache coherence is an integral part of shared-memory systems but is also widely
considered to be one of the most complex parts of such systems. Much prior work has …
considered to be one of the most complex parts of such systems. Much prior work has …
Accelerating cache coherence in manycore processor through silicon photonic chiplet
Cache coherence overhead in manycore systems is becoming prominent with the increase
of system scale. However, traditional electrical networks restrict the efficiency of cache …
of system scale. However, traditional electrical networks restrict the efficiency of cache …
The effects of granularity and adaptivity on private/shared classification for coherence
Classification of data into private and shared has proven to be a catalyst for techniques to
reduce coherence cost, since private data can be taken out of coherence and resources can …
reduce coherence cost, since private data can be taken out of coherence and resources can …
Leveraging Cache Coherence to Detect and Repair False Sharing On-the-fly
V Patel, S Biswas, M Chaudhuri - 2024 57th IEEE/ACM …, 2024 - ieeexplore.ieee.org
Performance bugs due to false sharing do not man-ifest as observable correctness errors,
and hence are challenging to detect and repair. Prior approaches aim to both detect and …
and hence are challenging to detect and repair. Prior approaches aim to both detect and …
Data placement across the cache hierarchy: Minimizing data movement with reuse-aware placement
A Sembrant, E Hagersten… - 2016 IEEE 34th …, 2016 - ieeexplore.ieee.org
Modern processors employ multiple levels of caching to address bandwidth, latency and
performance requirements. The behavior of these hierarchies is determined by their …
performance requirements. The behavior of these hierarchies is determined by their …
Comparison of significant issues in multicore cache coherence
AD Joshi, N Ramasubramanian - … international conference on …, 2015 - ieeexplore.ieee.org
Multicore processors are becoming dominant in todays computing environment. All
multicore processors are with cache coherent memory model. Users can have multiple …
multicore processors are with cache coherent memory model. Users can have multiple …