See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/332745421

Relational Databases Versus HBase: An Experimental Evaluation

Article in Advances in Science Technology and Engineering Systems Journal · April 2019
DOI: 10.25046/aj040249

CITATION READS
1 1,623

3 authors:

Zakaria Bousalem Inssaf El Guabassi

Université Sultan Moulay Slimane Chouaib Doukkali university, Faculty of Sciences
17 PUBLICATIONS 171 CITATIONS 22 PUBLICATIONS 223 CITATIONS

SEE PROFILE SEE PROFILE

Ilias Cherti
Université Hassan 1er
25 PUBLICATIONS 59 CITATIONS

SEE PROFILE

All content following this page was uploaded by Inssaf El Guabassi on 02 May 2019.

The user has requested enhancement of the downloaded file.

 Advances in Science, Technology and Engineering Systems Journal Vol. 4, No. 2, 395-401 (2019) ASTESJ
www.astesj.com ISSN: 2415-6698

Relational Databases Versus HBase: An Experimental Evaluation

Zakaria Bousalem*,1, Inssaf El Guabassi2, Ilias Cherti1
1
Faculty of Sciences and Technologies, Hassan 1st University, Settat, Morocco
2
Faculty of Sciences, Abdelmalek Essaadi University, Morocco

ARTICLE INFO ABSTRACT

Article history: Relational database management systems (RDBMS) have been imposed for more than
Received:21 February, 2019 three decades as a facto standard for data storage, management, and analysis. They have
Accepted:07 April, 2019 a good reputation by supporting ACID properties (Atomicity, Consistency, Isolation, and
Online: 26 April, 2019 Durability) and by adopting the SQL language which has become a standardized
language. However, despite their power, RDBMS have failed to meet the modern
Keywords:
application's requirements. That's why the need arises for new database management
Relational database
systems that support the manipulation of large amounts of data. NoSQL database systems
HBase
allow a flexible schema, whereas RDBMSs require a strictly defined schema. They
NoSQL
support horizontal scalability and prioritize data availability over consistency (BASE
Comparative study
properties) and have performance that remains good with scalability. In this paper, we
Benchmarking
present an experimental comparison between a relational database (MySQL) and a
YCSB
NoSQL database (HBase) in terms of runtime and latency in different scenarios using the
YCSB Framework.

1. Introduction Column Oriented databases and Graph databases. This

classification is due to the fact that each type of database arises
For more than three decades Relational databases has been in a specific context and based on different architectures [4].
the de-facto standard in the database management systems Comparing different models provides a clear vision for choosing
market thanks to its maturity[1], [2]. Nowadays, with a constant the most appropriate model for a given context. The purpose of
growth of data generated by modern web applications such as this article is to compare the relational model (MySQL) and the
social networks, e-commerce sites, and mobile applications; the NoSQL model (HBase)[5] in terms of runtime and latency in
management, querying and analysis data have become a real different scenarios using the YCSB Framework. We will
challenge for relational database management systems measure the latency of three cases of operations: 100% read
(RDBMS). Besides, these data are recorded in several formats operations, 100% update operations, and a mix of 50% reads and
(structured, semi-structured and unstructured), whereas the 50% updates with two scenarios. The first is increasing the
traditional database management systems based on a rigid number of records however the total number of operations
schema. These limitations of the relational model led the leaders remains fixed at 10000. The second is increasing the number of
of the internet such as Google, Amazon, eBay, Alibaba and operations while fixing the total number of records at 1 million
Facebook to develop a new model named NoSQL databases[3], records in order to reveal how the number of operations and
in order to overcome the weakness of relational database number of records affect the performance in terms of the latency
management systems towards the variety, the velocity and the metric and runtime for data loading. In order to make an
large volume of new data captured. "NoSQL" databases are not efficient approach for migration from Relational databases to
usually a replacement, but rather a complementary complement HBase database, we have started by a feasibility assessment[6],
to RDBMS and SQL. The NoSQL model is based on the CAP and in this paper, we have made an experimental comparison
theorem (Consistency Availability Partition Tolerance) as between relational databases and HBase database. The goal of
opposed to RDBMS based on ACID properties (Atomicity, this comparison is clearly identifying which case is better to
Coherence, Isolation, Durability). NoSQL databases migrate from relational to HBase.
management systems (DBMS) can be classified into four
categories: Key-Value databases, Document Oriented databases, The rest of the paper is structured as follows: In section 2 we
*
Zakaria Bousalem, Faculty of Sciences and Technologies, Hassan 1st
introduce the basic definitions starting with an introduction to the
University, Settat, Morocco, [email protected] NoSQL databases after we present the HBase database, so we
www.astesj.com 395
 Z. Bousalem et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 4, No. 2, 395-401 (2019)
will see what Databases Benchmarking is, then we provide a brief  Table: HBase was organizing data in tables.
presentation of YCSB Framework. In section 3 we introduce the  Row: Within tables, the data is organized in rows.
experimental strategy used in our paper. In section 4 we describe RowKey is the identifier for each row.
the experimental setup for evaluation. In Section 5 we present the  Column Family: In each row, data is grouped by
MySQL and HBase evaluation results. In Section 6, a summary "Column Families ". All rows have the same "Column
and general observations about the results of this evaluation are Families". The "Column Family" is set when the table is
provided. Finally, Section 7 concludes our paper. created in HBase.
 Column Qualifier: Access to data within a "Column
2. Basic definitions
Family" is done via the "column qualifier". It’s specified
2.1. NoSQL Databases at the data insertion phase.
 Cell: Cell is identified by the combination of the
NoSQL (Not Only SQL) is a broad category of next- "RowKey", the "Column Family" and the "Column
generation database management systems, as they are typically Qualifier". It’s Stores the values.
non-relational, distributed, open source, and support horizontal  Version: The values within a cell are versioned. The
scaling. Unlike relational databases, they can better respond to versions are identified by their timestamp.
big data problems. These database systems do not rely on a rigid
relational schema and the database can therefore grow without
constraint.
There are various classes of NoSQL DBMS [7]:
 Key / Value: These databases function as a key/value
associative array. This structure makes it a simple database
to set up and allows quick access to information. The value
can be a string or an object. It offers high scalability thanks
to schema-less approach. E.g. Riak, Azure Table Storage,
and Redis.
 Document-Oriented databases: These databases Figure 1: HBase model [12]
management systems are an extension of the key/value
databases. Document-oriented engines do not associate a key 2.3. Databases Benchmarking
with a value but with a schema-less document like JSON and
Database benchmarks (Performance evaluation by
XML. The flexibility of these databases makes them
polyvalent. E.g. MongoDB, Couchbase Server, and experimentation on a real system) [13] are an important tool for
OrientDB. database researchers, designers, and users. Its role is to generate
application-specific workloads and to test databases in order to
 Column-oriented databases: The data representation is assess the relative performance and ease the process of making
done by columns contrary to traditional DBMS. This comparisons between different database specifications. As
structure makes it easier to add a column to a table and mentioned by [14] the big data benchmarking process is
manage millions of columns. These databases are known for composed of five steps: Planning, Generating data, Generating
their ability to scale and to store a large volume of data. tests, Execution and Analysis, and evaluation Figure 2[14].
These DBMSs are mainly used in environments where it is
necessary to access many columns. They are especially
useful for streaming data and Real-time analytics. E.g.
HBase, Cassandra, and BigTable.
 Graph databases: Store data based on graph theory using Figure 2: Benchmarking process for big data systems [14]
graph structures (nodes, arcs, and properties). This storage There are many existing tools for Big Data benchmark [15]
model facilitates the representation of all highly connected
like BigBench [16], TPC-C[17] , TPC-E[18], TPC-H[19], TPC-
data, which is particularly well adapted to the social
networks data processing, fraud detection, and D[20], Bigdatabench [21] and YCSB [22]. In this paper, we are
recommendation engine[8]. E.g. AllegroGraph, Neo4j, and going to use Yahoo! Cloud Serving Benchmark (YCSB) because
FlockDB. is currently the most popular choice for benchmarking
performance of big data databases [23].
2.2. HBase
2.4. YCSB
HBase is a distributed database management system,
developed on top of the HDFS file system. It belongs to the YCSB benchmark is an extensible, modular benchmarking
column-oriented databases category. HBase is designed to tool, it was developed by Yahoo teams for the aim of measuring
provide real-time access to data stored on HDFS. It supports the performance of various storage solutions, with adapters for a
horizontal scalability which allows it to support extremely large variety database systems such as Relational databases (with
database tables[9]. It was based on "BigTable" DBMS [10]. JDBC driver), Big data databases(HBase, Mongo, Cassandra,
Redis, HyperTable, Couchbase, DynamoDB, Accumulo, etc ) and
As shown in Figure 1, the HBase data model is based on six others.
concepts [11]:
www.astesj.com 396
 Z. Bousalem et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 4, No. 2, 395-401 (2019)
As shown in Figure 3 YCSB client is composed of four running time of five workloads. According to their results, when
modules: the executor workload, client-threads module, performing reads, MongoDB showed good performance, unlike
Database interface module and statistics module. After HBase which showed good performance for updates.
generating the data to be loaded to the database by YCSB client,
Latency means the time of response can get user when
the executor workload will launch several client threads that
sending a request. It’s one of the essential metrics to evaluate
execute a series of operations (client-threads module) by using
databases performance [27]. Consequently, we chose in our
the (Database Interface Layout). The statistics module will
paper to compare MySQL and HBase in terms of runtime and
retrieve statistics from each operation and analyze them.
response time (latency) based on read and update operations
since they are the most used operations[24] while increasing the
number of records however the total number of operations
remains fixed at 10000. Then we will increase the number of
operations while fixing the total number of records at 1 million
records in order to reveal how the number of operations and
number of records affect the performance in terms of latency and
runtime for data loading.
4. Experimental setup
In order to perform our comparative study, we present the
Figure 3: YCSB client architecture [22] experimental setup for evaluation. The experiments were run
using a single physical machine with Ubuntu operating system,
YCSB benchmark varies the proportion of read, write,
YCSB benchmark, Cloudera Hadoop, Cloudera HBase, and,
update, insertion, and scan operations in a series of queries
MySQL. All specifications are listed in Table1.
named workloads. The YCSB distribution includes six
workloads: Table 1 Experimental specifications

 Workload A: A mixed workload with 50% of reads and CPU Intel® Xeon(R)
50% of writes CPU E5504 @ 2.00GHz × 8
 Workload B: A mixed workload with 95% of reads and Memory 16 GB
5% of writes. Hard disk 237 GB SSD
 Workload C: A workload of 100% read Operating system Ubuntu 14.04 (64-bit)
 Workload D: A mixed workload with 95% of reads and Java version 1.8.0_16
5% of inserts. YCSB version 0.14
CDH version 5.14.1
 Workload E: A mixed workload with 95% of scans and
Cloudera HBase version 1.2.0
5% of inserts.
Cloudera Hadoop version 2.6.0
 Workload F: Read-modify-write: A mixed workload
MySQL 5.6.26
with 50% of reads and 50% of read-modify-writes
The main focus of this study is to evaluate read and update
3. Experimental strategy
operations since they are the most used operations [24].
Much work on the potential of comparing database Therefore this comparison mainly consists of three workloads
performances by YCSB has been carried out [24]–[26]. namely A and B included in the YCSB project and we create
Abramova et al [24] compare five NoSQL databases (Redis, new workload G proposed by [24] to evaluate the Update Only
Cassandra, HBase, MongoDB, and OrientDB) in terms of their case. Table 2 shows the tested workloads:
capabilities, based on read and update operations. They affirm Table 2 Used workloads
that MongoDB, Redis, and OrientDB are better for reads,
Cassandra and HBase are optimized for updates. Yassien and Workload Operations
Desouky [26] compare MySQL, MongoDB, and HBase by using Workload A 50% of reads and 50% of writes
YCSB for the aim to study the effect of varying the operation Workload C 100% read: Read Only
and thread count with respect to runtime, throughput, and Workload G 100% update : Update Only
latency. The authors state that each database performs at its best The dataset used in this databases benchmarking is generated
in different circumstances. They recommend HBase to use for by YCSB data generator which is a part of YCSB client. The
the applications that require the high update and insert dataset records are composed of 10 fields. Each field is filled by
operations, MySQL for the applications whose perform mostly a random string with 100 bytes which give 1 KB per record. The
reads operations and MongoDB for the applications that require ‘YCSB_KEY’ is the primary key for each row[22]. Table 3
both adequate read and write performance. Matallah et al [25] shows the YCSB dataset structure.
compare MongoDB and HBase in order to evaluate loading and

www.astesj.com 397
 Z. Bousalem et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 4, No. 2, 395-401 (2019)
Table 3 YCSB Dataset structure

YCSB_KEY FIELD1 FIELD2 FIELD3 FIELD4 FIELD5 FIELD6 FIELD7 FIELD8 FIELD9
Row 1
Row 2
......
Row N

5. Experimental results 5.2. Increasing the number of records

We performed three tests. The first is loading data, the second 5.2.1. Workload A
is running workloads while increasing the number of records and  Runtime: As illustrated in Figure 6 as the size of data
fix the number of operations at 10000 and the last is running increases the runtime of MySQL for data loading
workloads while increasing the number of operations and set the increases. MySQL exhibits a slight steady increase in
number of records to 1 million. runtime, unlike HBase that shows a slight decline and
5.1. Loading data increase thereafter. HBase has the lowest runtime.

 Runtime (less is better): As shown in Figure 4 as the size

of data increases the runtime of loading data for MySQL
and HBase increases, HBase exhibited an immense
increase but MySQL shows starting from 100000 records
a dramatic increase. Additionally, HBase has the lowest
runtime. In the first test (Record number=1000) MySQL
and HBase have almost identical runtime. As the record
number increases, the runtime for MySQL to load data
ranges from 2 times slower than HBase for the second
test(Record number=10000), to more than 4 times slower
for MySQL for the third test and more than 5 times for the
fourth test. Figure 6: Workload A Runtime
 Read latency (less is better): HBase shows a slight
decline and increases thereafter, unlike MySQL that
exhibits a slight steady increase as shown in Figure 7.
MySQL has the shortest read latency.

Figure 4: Loading data Runtime

 Insert latency (less is better): decreases as the size of data
increases for HBase. Unlike MySQL that shows a steadiness
initially and then it exhibited a decline and increase
thereafter. As shown in Figure 5 HBase has the shortest
insert latency. Figure 7: Workload A Read Latency

Figure 5: Loading data Insert latency Figure 8: Workload A Update Latency

www.astesj.com 398
 Z. Bousalem et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 4, No. 2, 395-401 (2019)
 Update latency (less is better): As shown in Figure 8,
like runtime in update latency MySQL exhibits a slight
steady increase in runtime, unlike HBase that shows a
slight decline and increase thereafter. HBase has the
lowest update latency.

5.2.2. Workload C
As illustrated in Figure 9 and Figure 10 HBase exhibits a
slightly decline initially, it shows an alternating increase and
decline thereafter in terms of runtime and read latency, unlike
Figure 12: Workload G Latency
MySQL that shows a steadiness initially, then it exhibited a slight
increase after reaching 100000 records. MySQL has the shortest
5.3. Increasing the number of operations
run time and read latency.
5.3.1. Workload A

 Runtime: As shown in Figure 13 as the number of

operations increases the runtime of MySQL and HBase
increases, HBase and MySQL show an immense increase
but MySQL exhibits starting from 100000 operations a
dramatic increase. HBase has the lowest runtime.

Figure 9: Workload C Runtime

Figure 13: Workload A Runtime

 Read latency (less is better): HBase and MySQL show
an immense decline until reaching 10000 records, then
exhibit slightly decline as shown in Figure 14. MySQL
has the shortest read latency.

Figure 10: Workload C Read Latency

5.2.3. Workload G

As illustrated in Figure 11 and Figure 12 the HBase exhibits a

steadiness both for runtime and update latency, unlike MySQL
that shows a slight increase. HBase has the lowest value of
runtime and read latency.

Figure 14: Workload A Read Latency

Figure 11: Workload G Runtime

Figure 15: Workload A Update Latency
www.astesj.com 399
 Z. Bousalem et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 4, No. 2, 395-401 (2019)
 Update latency (less is better): As illustrated in Figure  Update latency (less is better): HBase shows an
15 as the number of operations increases the update immense decrease, unlike MySQL that exhibits a steady
latency of HBase decline. MySQL exhibits a slightly increase as shown in Figure 19. HBase has the shortest
decline initially, it shows an alternating increase and a update latency.
steadiness thereafter in terms of update latency. HBase
has the lowest update latency.
5.3.2. Workload C

 Runtime: As shown in Figure 16 as the number of

operations increases the runtime of MySQL and HBase
increases, HBase and MySQL show an immense growth
but starting from 100000 operations they exhibit a
dramatic increase. MySQL has the lowest runtime.

Figure 19: Workload G Update Latency

6. General observations

According to our experimental results, it can be observed

that MySQL runtime is higher in all scenarios for data loading
and HBase performed far better compared to MySQL.
Concerning read/write latencies, it can be stated that MySQL’s
latency is lower for read operations and HBase’s latency is lower
Figure 16: Workload C Runtime for write operations. In terms of the running workloads runtime;
HBase beats the competition in all cases except for the read-only
 Read latency(less is better): HBase shows an immense
workload. Also from our experimental results it can be stated
decrease, unlike MySQL that exhibits a steady decline as
that on the one hand the increasing the number of records seems
shown in Figure 17. MySQL has the shortest read latency.
to have mediocre effects on read latency for MySQL and HBase,
no consequences in term of read latency and great impact with
respect to runtime data loading (Taken into account that number
of records tested was not of a real large size). On the other hand,
increasing the number of operations seems to have a significant
impact on read and write latency for MySQL and HBase, and
immense effects on the running workloads runtime.
Consequently, we believe that HBase outperforms MySQL on
I/O bound (‘write’) operations but lagged behind in bound
(‘read’) operations with respect to runtime and latency metrics.
HBase exhibits good performance in update operations thanks to
Figure 17: Workload C Read Latency using the log files and cache memories to store all transactions
5.3.3. Workload G and then write only the log files on disk which reduce the
input/output operations [25], contrary to MySQL that stores data
 Runtime: As shown in Figure 18 as the number of directly on disk. Additionally, HBase lagged behind in reads
operations increases the runtime of MySQL and HBase capabilities due to comparing all copies by HBase before
increases, HBase and MySQL show an immense increase running a read operation in order to return the most recent copy,
but MySQL exhibits starting from 100000 operations a which affects database performance [25]. So, according to our
dramatic increase. HBase has the lowest runtime. experimental results, we can say that is better to migrate from
Relational databases (MySQL) to HBase in case of the
applications that require a heavy update, most update and high
insert operations like session store in order to record recent
actions.

7. Conclusion and future work

In this paper, we present an experimental comparison

between a relational database (MySQL) and a NoSQL database
(HBase) with respect to runtime and latency in different
Figure 18: Workload G Runtime
scenarios using the YCSB Framework. Based on the above
www.astesj.com 400
 Z. Bousalem et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 4, No. 2, 395-401 (2019)
results we can deduce that HBase performed far better compared [26] A. W. Yassien and A. F. Desouky, “RDBMS, NoSQL, Hadoop: A
Performance-Based Empirical Analysis,” in Proceedings of the 2nd Africa
to MySQL in data loading because MySQL runtime is higher in and Middle East Conference on Software Engineering, 2016, pp. 52–59.
all scenarios for this kind of operation. Additionally, we have [27] X. Tian, R. Han, L. Wang, G. Lu, and J. Zhan, “Latency critical big data
found that HBase outperforms MySQL on I/O bound (‘write’) computing in finance,” J. Finance Data Sci., vol. 1, no. 1, pp. 33–41, 2015.
operations but lagged behind in bound (‘read’) operations with
respect to runtime and latency metrics. In perspective, we
envisage to compare MySQL and HBase in terms of database
performance of the aggregate functions and also pass to higher
scales by using a very large database and performing the
evaluation in a really distributed and parallel environment.
References
[1] Z. Bousalem and I. Cherti, “XMap: A Novel Approach to Store and
Retrieve XML Document in Relational Databases.,” JSW, vol. 10, no. 12,
pp. 1389–1401, 2015.
[2] Z. Bousalem, I. El Guabassi, and I. Cherti, “Toward Adaptive and Reusable
Learning Content Using XML Dynamic Labeling Schemes and Relational
Databases,” in Advanced Intelligent Systems for Sustainable Development
(AI2SD’2018), 2019, pp. 787–799.
[3] D. J. Abadi, “Data management in the cloud: Limitations and
opportunities.,” IEEE Data Eng Bull, vol. 32, no. 1, pp. 3–12, 2009.
[4] J. R. Lourenço, B. Cabral, P. Carreiro, M. Vieira, and J. Bernardino,
“Choosing the right NoSQL database for the job: a quality attribute
evaluation,” J. Big Data, vol. 2, no. 1, p. 18, 2015.
[5] L. George, HBase: the definitive guide: random access to your planet-size
data. O’Reilly Media, Inc., 2011.
[6] Z. Bousalem, I. Cherti, and G. Zhao, “Migration from Relational Databases
to HBase: A Feasibility Assessment,” in International Conference on
Advanced Information Technology, Services and Systems, 2017, pp. 383–
395.
[7] A. B. M. Moniruzzaman and S. A. Hossain, “NoSQL Database: New Era
of Databases for Big data Analytics - Classification, Characteristics and
Comparison,” ArXiv13070191 Cs, Jun. 2013.
[8] J. Webber and I. Robinson, “The Top 5 Use Cases of Graph Databases,”
Neo Technol., 2015.
[9] N. Dimiduk, A. Khurana, M. H. Ryan, and M. Stack, HBase in action.
Manning Shelter Island, 2013.
[10] F. Chang et al., “Bigtable: A distributed storage system for structured data,”
ACM Trans. Comput. Syst. TOCS, vol. 26, no. 2, p. 4, 2008.
[11] A. Khurana, “Introduction to HBase schema design,” White Pap. Cloudera,
2012.
[12] Blandine Larbret, “Hadoop Hbase - Introduction,” 09:09:39 UTC.
[13] J. Darmont, “Data-Centric Benchmarking,” in Encyclopedia of Information
Science and Technology, Fourth Edition, IGI Global, 2018, pp. 1772–1782.
[14] R. Han, X. Lu, and J. Xu, “On big data benchmarking,” in Workshop on
Big Data Benchmarks, Performance Optimization, and Emerging
Hardware, 2014, pp. 3–18.
[15] R. Han, L. K. John, and J. Zhan, “Benchmarking big data systems: A
review,” IEEE Trans. Serv. Comput., vol. 11, no. 3, 2018.
[16] A. Ghazal et al., “BigBench: towards an industry standard benchmark for
big data analytics,” in Proceedings of the 2013 ACM SIGMOD
international conference on Management of data, 2013, pp. 1197–1208.
[17] “Tpc-c.” [Online]. Available: http://www.tpc.org/tpcc/.
[18] “Tpc-e.” [Online]. Available: http://www.tpc.org/tpce/.
[19] “Tpc-h.” [Online]. Available: http://www.tpc.org/tpch/.
[20] “Tpc-ds.” [Online]. Available: http://www.tpc.org/tpcds/.
[21] L. Wang et al., “Bigdatabench: A big data benchmark suite from internet
services,” in High Performance Computer Architecture (HPCA), 2014
IEEE 20th International Symposium on, 2014, pp. 488–499.
[22] B. F. Cooper, A. Silberstein, E. Tam, R. Ramakrishnan, and R. Sears,
“Benchmarking cloud serving systems with YCSB,” in Proceedings of the
1st ACM symposium on Cloud computing, 2010, pp. 143–154.
[23] D. Bermbach, E. Wittern, and S. Tai, “Getting Started in Cloud Service
Benchmarking,” in Cloud Service Benchmarking, Springer, 2017, pp. 151–
153.
[24] V. Abramova, J. Bernardino, and P. Furtado, “Which nosql database? a
performance overview,” Open J. Databases OJDB, vol. 1, no. 2, pp. 17–24,
2014.
[25] H. Matallah, G. Belalem, and K. Bouamrane, “Experimental comparative
study of NoSQL databases: HBASE versus MongoDB by YCSB,” Comput
Syst Sci Eng, vol. 32, no. 4, pp. 307–317, 2017.
www.astesj.com 401

View publication stats