Experimental dataset for the Performance Evaluation for Geographically Distributed Blockchain-based Services in a Cloud Computing Environment

Citation Author(s):
TaeYoung
Kim
Sungkyunkwan University
Hyung-Jong
Kim
Seoul Women's University
Submitted by:
Kim TaeYoung
Last updated:
Tue, 01/18/2022 - 02:13
DOI:
10.21227/b7mg-yb75
Data Format:
*.txt; *.json
License:
Creative Commons Attribution
1061 Views
Categories:
Cloud Computing
Keywords:
Performance evaluation, Blockchain, Blockchain job workload data, cloud computing, Cloud computing dataset, AWS EC2
0
0 ratings - Please login to submit your rating.
ACCESS DATASET CITE SHARE/EMBED

Abstract 

This dataset is an experimental result of the paper “Performance Evaluation for Geographically Distributed Blockchain-based Services in a Cloud Computing Environment”. The Geographically Distributed Cloud Performance Evaluation Ambassador (GDCPEA) is deployed on each Go Ethereum (Geth) node to measure the elapsed time from the start to the end of the Geth main operations. When GDCPEA requests the GDCPE Log Server in the same region to record a log, the GDCPE Log Server records a timestamp log including the timestamp when the request was received, the name of the Geth node that sent the request, the name of the Geth operation to be measured, and so on. We conducted this experiment in AWS cloud computing services and deployed GDCPEA and GDCPE Log Server in up to three AWS regions (Seoul, Ohio, Singapore) depending on the purpose of the experiment. This dataset contains timestamp logs from each GDCPE Log Server, block and transaction information from the blockchain network, and resource usage data from each AWS EC2 instance. We attached the results of analyzing the performance changes when the number of nodes and clients of the blockchain network increases using these data to our paper.

Instructions: 

1. Dataset title: Experimental dataset for the Performance Evaluation for Geographically Distributed Blockchain-based Services in a Cloud Computing Environment

 

2. File name structure:

- [Number of nodes & Number of clients] data type (round).txt

- [Usage data type] GethA/B_data type.txt

 

3. Attributes

① Number of nodes = Number of nodes used in experiment  (3 ~ 5)

② Number of clients = Number of clients used in experiment (10, 30, 40, 50, 60, 80, 100, 120, 160, 200)

③ data type = Name of the data type

  (1) block_info : Information on each mined block

  (2) tx_info: Information on each verified transaction

  (3) GDPEA_seoul : Performance evaluation log requested by GDPEA in seoul region

  (4) GDPEA_ohio : Performance evaluation log requested by GDPEA in ohio region

  (5) cpu0/1 : CPU usage data provided by AWS CloudWatch Service

  (6) write time/writing operation/written bytes : Disk usage data provided by AWS CloudWatch Service

④ Usage data type = Name of the usage data type (CPU Usage, Disk Usage)

⑤ round : Number of experiment rounds (1 ~ 3)

 

4. File content format

  (1) block_info : block number, TxRoot hash

  (2) tx_info: block number, Tx number, Tx hash

  (3) GDPEA_seoul/ohio : timestamp, Geth nodeID, action, Tx sender, TxRoot hash, block number

  (*) Cloud Environment Resource Usage Data is a json format data provided by Amazon CloudWatch which consists of Timestamp, Statistical value, and Unit. This data is generated from the collectd plugin installed on each EC2 instance and is collected every 60 seconds.

Dataset Files

LOGIN TO ACCESS DATASET FILES

 

QUESTIONS?