MPEC: Distributed Matrix Multiplication Performance Modeling on a Scale-out Cloud Environment for Data Mining Jobs
Many data mining workloads are being analyzed in large-scale distributed cloud computing environments which provide nearly infinite resources with diverse hardware configurations. To maintain cost-efficiency in such environments, understanding the characteristics and estimating the overheads of a di...
Lưu vào:
| Tác giả chính: | |
|---|---|
| Đồng tác giả: | |
| Định dạng: | BB |
| Ngôn ngữ: | English |
| Thông tin xuất bản: |
IEEE Xplore
2020
|
| Chủ đề: | |
| Truy cập trực tuyến: | http://tailieuso.tlu.edu.vn/handle/DHTL/9968 |
| Từ khóa: |
Thêm từ khóa bạn đọc
Không có từ khóa, Hãy là người đầu tiên gắn từ khóa cho biểu ghi này!
|
| id |
oai:localhost:DHTL-9968 |
|---|---|
| record_format |
dspace |
| spelling |
oai:localhost:DHTL-99682020-12-23T07:44:47Z MPEC: Distributed Matrix Multiplication Performance Modeling on a Scale-out Cloud Environment for Data Mining Jobs Kim, Jeongchul Son, Myungjun Lee, Kyungyong Big data analytics Distributed matrix multiplication Performance modeling Cloud computing Many data mining workloads are being analyzed in large-scale distributed cloud computing environments which provide nearly infinite resources with diverse hardware configurations. To maintain cost-efficiency in such environments, understanding the characteristics and estimating the overheads of a distributed matrix multiplication task that is a core computation kernel in many machine learning algorithms are essential. This study aims to propose a Matrix Multiplication Performance Estimator on Cloud (MPEC) algorithm. The proposed algorithm predicts the latency incurred when executing distributed matrix multiplication tasks of various input sizes and shapes with diverse instance types and a different number of worker nodes on cloud computing environments. To achieve this goal, we first analyze the characteristics of distributed matrix multiplication tasks. With characteristics generated from qualitative analysis, we propose to apply an ensemble of non-linear regression algorithm to predict the execution time of arbitrary matrix multiplication tasks. Thorough experimental results reveal that the proposed algorithm demonstrates higher accuracy than a state-of-the-art machine learning task performance estimation engine, Ernest, by decreasing the Mean Absolute Percentage Error (MAPE) in half. https://doi.org/10.1109/TCC.2019.2950400 2020-12-23T07:43:47Z 2020-12-23T07:43:47Z 2019 BB http://tailieuso.tlu.edu.vn/handle/DHTL/9968 en IIEEE Transactions on Cloud Computing, (2019), pp 18 application/pdf IEEE Xplore |
| institution |
Trường Đại học Thủy Lợi |
| collection |
DSpace |
| language |
English |
| topic |
Big data analytics Distributed matrix multiplication Performance modeling Cloud computing |
| spellingShingle |
Big data analytics Distributed matrix multiplication Performance modeling Cloud computing Kim, Jeongchul MPEC: Distributed Matrix Multiplication Performance Modeling on a Scale-out Cloud Environment for Data Mining Jobs |
| description |
Many data mining workloads are being analyzed in large-scale distributed cloud computing environments which provide nearly infinite resources with diverse hardware configurations. To maintain cost-efficiency in such environments, understanding the characteristics and estimating the overheads of a distributed matrix multiplication task that is a core computation kernel in many machine learning algorithms are essential. This study aims to propose a Matrix Multiplication Performance Estimator on Cloud (MPEC) algorithm. The proposed algorithm predicts the latency incurred when executing distributed matrix multiplication tasks of various input sizes and shapes with diverse instance types and a different number of worker nodes on cloud computing environments. To achieve this goal, we first analyze the characteristics of distributed matrix multiplication tasks. With characteristics generated from qualitative analysis, we propose to apply an ensemble of non-linear regression algorithm to predict the execution time of arbitrary matrix multiplication tasks. Thorough experimental results reveal that the proposed algorithm demonstrates higher accuracy than a state-of-the-art machine learning task performance estimation engine, Ernest, by decreasing the Mean Absolute Percentage Error (MAPE) in half. |
| author2 |
Son, Myungjun |
| author_facet |
Son, Myungjun Kim, Jeongchul |
| format |
BB |
| author |
Kim, Jeongchul |
| author_sort |
Kim, Jeongchul |
| title |
MPEC: Distributed Matrix Multiplication Performance Modeling on a Scale-out Cloud Environment for Data Mining Jobs |
| title_short |
MPEC: Distributed Matrix Multiplication Performance Modeling on a Scale-out Cloud Environment for Data Mining Jobs |
| title_full |
MPEC: Distributed Matrix Multiplication Performance Modeling on a Scale-out Cloud Environment for Data Mining Jobs |
| title_fullStr |
MPEC: Distributed Matrix Multiplication Performance Modeling on a Scale-out Cloud Environment for Data Mining Jobs |
| title_full_unstemmed |
MPEC: Distributed Matrix Multiplication Performance Modeling on a Scale-out Cloud Environment for Data Mining Jobs |
| title_sort |
mpec: distributed matrix multiplication performance modeling on a scale-out cloud environment for data mining jobs |
| publisher |
IEEE Xplore |
| publishDate |
2020 |
| url |
http://tailieuso.tlu.edu.vn/handle/DHTL/9968 |
| work_keys_str_mv |
AT kimjeongchul mpecdistributedmatrixmultiplicationperformancemodelingonascaleoutcloudenvironmentfordataminingjobs |
| _version_ |
1768589931608801280 |