Journal of Environmental Treatment Techniques
2020, Volume 8, Issue 1, Pages: 787-793
.2 Characteristics of NoSQL database
NoSQL database is available in four generic types:
document-based, column-based, key-value, and graph .
NoSQL database has an advantage over relational database
due to its “flexi-schema”. The “flexi-schema” behaviour
allows different structures of records to be stored within the
same table . For example, in a document-based NoSQL
such as MongoDB, documents (record) within the same
collection (table) are allowed to contain different numbers of
fields. The “shared nothing architecture” of NoSQL applies
local storage pool that allows faster data access by adding
number of data nodes. NoSQL database has such a high
elasticity that replicates data to newly-added data nodes .
Eventually consistency, data can be read from replicas of
other data machine if a machine is down .
Figure 1: MongoDB document 
c) Supporting features: Even within the same type
of NoSQL database, other features such as support on
query language and CAP features support are different
11]. The challenges mentioned above are based on the
.3. Challenges of data migration in NoSQL databases
Three challenges commonly arise in data migration :
a) interruption of business operation; (b) loss of data and
features of NoSQL databases. In the perspective of
data quality, data migration may face the challenges in
maintaining the quality of data migrated. Table 2
summarizes the challenges of data migration from the
perspective of data quality.
degradation of data consistency and (c) effort and cost
required for data migration. In NoSQL database, data
migration process faces some challenges related to the
common challenges sated above, which are caused by the
characteristics of the NoSQL database.
Table 2: Data quality challenges of data migration .
a) Heterogeneous storage paradigm: Each type of
NoSQL database implements different ways to store data.
Different storage paradigms have specific rules and format
in storing the data. Table 1 summarizes the storage
paradigms of NoSQL databases. The table indicates that re-
formatting or restructuring of data is required in order to map
the data to the targeted database storage structure.
Therefore, qualities of data such as completeness,
consistency, and correctness are concerned when data are
being restructured or reformatted. The degradation of data
quality will lead to higher cost of recovery and data quality
The old and new database may have
different fields. Some fields in the new
database may not exist in the old one. The
NULL value is used to represent the non-
existence of data, which is critical for
migration of data between different
When data is migrated through manual
approach, especially keyed by human,
accuracy of data is not guaranteed.
Some data may lose as the result of system
Existing data has the problem where same
word is used for different definitions.
Therefore, it needs further clarification on
the value of data transferred. This problem
affects data consistency.
Table 1: NoSQL database storage paradigms.
Allowing embedded of key value in
document; allowing search based on
both key and value 
Storing data in distributed, multiple
dimensional map; having mixed
According to the challenges discussed in this section, it is
clear that data migration requires not only to ensure the data
can be migrated, but also concern the quality of migrated data.
For NoSQL databases, the challenges discussed in this
section have significant effects on data quality. For example,
the flexibility of schema in NoSQL may cause missing data
in some fields of target database.
row/column storage 
Storing data in byte-array; assessing
data through key-value hash table
(each key points to a specific datum)
Storing data in nodes; connecting data
by edge (edge represents the
relationships between nodes); using
pointer to point to another nodes 
3 Related Work
The authors in  attempted to overcome the challenge
of different data formats between NoSQL databases. To this
end, they proposed an approach of migrating data between
different types of NoSQL databases by converting the
existing data into an intermediate format to be converted later
again to the format required by the destinated database. The
approach migrated data between column-based NoSQL
b) Flexibility in schema structure: The “flexi-schema” of
NoSQL allows more flexibility in storing data . For
example, in MongoDB, documents in the same collection
may have different numbers of fields. Figures 1 and 2 
show the examples of different schema in documents.