Proceedings of the 28th ACM International Conference on Information and Knowledge Management 2019
DOI: 10.1145/3357384.3358015
|View full text |Cite
|
Sign up to set email alerts
|

Deploying Hash Tables on Die-Stacked High Bandwidth Memory

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
4
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
3
3

Relationship

0
6

Authors

Journals

citations
Cited by 7 publications
(4 citation statements)
references
References 19 publications
0
4
0
Order By: Relevance
“…Second, data processing with HBM/HMC. Previous work [16], [17], [18], [19], [20], [21], [22], [23] employs HBM to accelerate their applications, e.g., hash table deep learning and streaming, by leveraging the high memory bandwidth provided by Intel Knights Landing (KNL)s HBM [24]. In contrast, we benchmark the performance of HBM on the Xilinx FPGA.…”
Section: Related Workmentioning
confidence: 99%
“…Second, data processing with HBM/HMC. Previous work [16], [17], [18], [19], [20], [21], [22], [23] employs HBM to accelerate their applications, e.g., hash table deep learning and streaming, by leveraging the high memory bandwidth provided by Intel Knights Landing (KNL)s HBM [24]. In contrast, we benchmark the performance of HBM on the Xilinx FPGA.…”
Section: Related Workmentioning
confidence: 99%
“…Multicore processors also take advantage of HBM, such as Intel's Knights Landing, NVIDIA's Titan V, and Google's TPU. Recent research in this area has focused on demonstrating the utility of HBM data-intensive processing issues, such as hash tables [38], graph processing [39], and stream processing [40]. When it comes to the process of expediting text search, there are two obstacles related to accessing the external memory:…”
Section: High Bandwidth Memorymentioning
confidence: 99%
“…KNL being an x86 many-core architecture offers easy portability for existing codebases and allows rapid testing of HBM-related ideas. Cheng et al [41] focus on optimizing NUMA placement of hash tables on KNL to increase the utilization of the HBM and provide simulation results for hash join. Pohl et al [42] focus on using the HBM on KNL for joins and find that the mode where HBM is directly addressed as opposed to the cache-mode results in the highest performance.…”
Section: Stochastic Gradient Descent (Sgd)mentioning
confidence: 99%