International Journal of Computer Applications (0975 – 8887)

Volume 182 – No. 48, April 2019

A Comparative Analysis of Symmetric Algorithms in

Cloud Computing: A Survey

Enas Elgeldawi Maha Mahrous Awny Sayed

Computer Science Department Computer Science Department Computer Science Department
Faculty of Science Minia University Faculty of Science
Minia University, Egypt Minia, Egypt Minia University, Egypt

ABSTRACT credibility.Cryptography is the key concept used to handle

Introducing Cloud computing to the globe has changed many data security concerns in cloud computing. Cryptography is to
conceptual and infrastructural bases for today’s and convert data into unreadable form during storage and
tomorrow’s computing. It has made the global thinking transmission. The unreadable form of data is known as cipher
migrates rapidly towards cloud based architecture. Clouds text. When data is received, the cipher text is converted back
bring out a variety of benefits including computing resources into its original form which is known as plain text.
configurability, cost controllability, sustainability, mobility Conversion of plain text to cipher text is known as encryption
and service flexibility. However, the new concepts that clouds and the reverse process is known as decryption. Encryption
introduce such as outsourcing, multi-tenancy, and resource takes place at sender’s end whereas decryption takes place at
sharing create new challenges and raise a broad range of receiver’s end. In section 2, we frame the way cryptographic
security and privacy issues. Cryptography is the art-of-science algorithms is classified. Section 3, explains Feistel network
of protecting data privacy by converting it to unreadable which is the major scheme most symmetric ciphers are based
format using standard mathematical techniques. This paper on. A detailed study for each algorithm is given in Section 4.
provides a comprehensive study for eight of the most common Computational analysis is discussed in Section 5. And finally
symmetric cryptographic algorithms, namely, DES, 3DES, the conclusion key notes are presented in Section 6.
Blowfish, Twofish, RC2, RC5, RC6 and AES. A comparative
analysis based on the structure of the algorithm, encryption
2. CLASSIFICATION OF
and decryption times, throughput and memory utilization has CRYPTOGRAPHIC ALGORITHMS
been performed to examine the performance of each Cryptography is the art of science of using mathematical
algorithm. techniques to convert plain text (P) into an unreadable cipher
text (C) according to some encryption algorithm that uses a
General Terms numeric or alpha numeric parameter called key (K1). and
Cloud Computing, Cryptography, Symmetric Algorithm similarly, reconvert the cipher text back to plain text via a
decryption algorithm which uses a decryption key (K2). This
Keywords can be interpreted as Cipher text C = E (P, K1) and Plain text
Cloud Computing, DES, 3DES, Blowfish, Twofish, RC2, P = D (C, K2). This is indicated by Figure 1.
RC5, RC6, AES
Sender Receiver
1. INTRODUCTION
Cloud computing is the delivery of computing services over Encryption Decryption
the Internet. When you share your latest news with your Transmission line
friends over Facebook, or upload your photos to Instagram, C = E(P, K1) P = D(C, K2)
you are using “Cloud Computing”. Instead of storing data on
your local system or downloading applications you need, you Figure 1: Encryption and Decryption Processes
use a service over the Internet at another location, to store
There are many Cryptographic algorithms in use for cloud
your information or use its applications. The cloud computing
computing, they can be classified as follows:
model allows individuals and organizations to use software
and hardware that are managed by a third party at remote 2.1 Symmetric Algorithms
locations to access information and to use a shared pool of Symmetric Algorithms (also known as secret key algorithm)
resources. use a single key for both encryption and decryption, this key
Cloud Computing has been defined by the National Institute is known to both the sender and receiver. Symmetric
of Standards and Technology (NIST) [1] as: “Cloud algorithms can be subdivided into: Block Cipher and Stream
computing is a model for enabling convenient, on-demand Cipher. The major difference between block cipher and
network access to a shared pool of configurable computing stream cipher is that block cipher takes a message and break it
resources (e.g., networks, servers, storage, applications, and into a fixed size of blocks and converts one block of the
services) that can be rapidly provisioned and released with message at a time [2]. On the other hand, a stream cipher
minimal management effort or service provider interaction.” algorithm typically encrypts and decrypts the text by
With the introduction of numerous cloud based services, the converting one byte of the text at a time.
possibility that sensitive information stored in remote servers
could be exposed to unwanted parties is compromised. If
security is not consistent and robust enough, the advantages
that cloud computing has to offer will have questionable

7
 International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 48, April 2019

2.2 Asymmetric Algorithms yielding F(Ri, Ki)

In asymmetric algorithms, encryption and decryption are b. The XOR of F(Ri, Ki) and the left half L is computed
performed using two different but mathematically related yielding Li ⊕ F(Ri, Ki)
keys. a public key and a private key. The public key may be
freely distributed, while its paired private key must remain c. The two sides are swapped to set the R and L for the next
secret. The public key is typically used for encryption, while round, that is Li+1 = Ri and Ri+1 = Li ⊕ F(Ri, Ki)
the private or secret key is used for decryption.
Ln+1 and Rn+1 are recombined to construct the cipher text.
Figure 2 shows the classification of cryptographic algorithms
along with examples of each classification and the year they Plain Text Master Key K
are created.
L0 R0
Cryptography

XOR F K0

Symmetric Asymmetric

L1 R1

Block Cipher Stream Cipher XOR F K1

DES 1977
3DES 1998 RC4 1994 RSA 1977
Blowfish 1993 A5/1 1987 DSA 1991
Twofish 1999 A5/2 1989 Diffie-Hellman 1976
RC2 1987 Pike 1994 El-Gamal 1985
Wake 1993 Ln Rn
RC5 1994 Elliptic curve 1985
RC6 1998 Sober 1997
AES 2001 XOR F Kn

Figure 2: Cryptographic Algorithms Classification

3. FEISTEL BLOCK CIPHER Ln+1 Rn+1

Most of the major symmetric ciphers are based on a cipher
known as the Feistel block cipher or Feistel network. This was
a block cipher developed by the IBM cryptography researcher Cipher Text
Horst Feistel in the early 70’s [3]. It is considered as a general
approach to build a block cipher. Such a network divides the Figure 3: General Structure of Feistel network
plain text into fixed size blocks then takes a block of the
plaintext and the key as inputs, and applies several iterations 4. SYMMETRIC ALGORITHMS ON
called rounds to generate the cipher text. Each round applies a THE SPOT
function (F) that consists of two main operations, substitution As mentioned in the previous section, most of the major
(or S-box) and permutation (or P-box). Substitution is a symmetric ciphers are based on Feistel block. In this section
mapping of one value to another whereas permutation is a bit- we review the major symmetric algorithms in used today,
shuffling process or simply a reordering of the bit positions namely, DES, 3DES, Blowfish, Towfish, RC2, RC5, RC6,
for each of the inputs. The block size, key size, the function F AES.
and number of rounds vary from one cipher algorithm to
another. Generally, the more rounds there are, and the larger 4.1 DES
the key size is, the more secure a cipher algorithm will be. The Data Encryption Standard (DES) is one of the most
Most symmetric encryption schemes today are based on this widely used and publicly available cryptographic algorithms.
structure. It was developed by IBM in the 1977 and it was the initial
The Feistel structure starts by splitting a block of plain text encryption standard to be recommended by the National
into two equal right (R) and left (L) halves. The function (F) Institute of Standards and Technology (NIST) [4]. DES is a
takes the right half and a sub key from the original master key block Cipher which is designed to encrypt and decrypt blocks
as inputs, the output of F is then XORed with the left half (L). of data consisting of 64 bits by using a 64-bit key. Although
Then the two sides are swapped to start another round. The the input key for DES is 64 bits long, the actual key used is
structure of Feistel algorithm is shown in Figure 3 and applies only 56 bits in length. The least significant bit in each byte is
the following steps. a parity bit. These parity bits are ignored, so only the seven
most significant bits of each byte are used, resulting in a key
1. The plain text is split into two halves L0 and R0. length of 56 bits. The algorithm goes through 16 Feistel
rounds. The algorithm transforms 64-bit input in a series of
2. For each round i = 1, 2, …, n
steps into a 64-bit output. The same steps, with the same key
a. The function F is applied to the right half R and ith sub key are used for decryption. Many attacks recorded the

8
 International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 48, April 2019

weaknesses of DES, which made it an insecure block cipher. Plain Text

Despite the growing concerns about its vulnerability, DES is
32 bits 32 bits
still widely used by financial services and other industries
worldwide for data protection. L1 R1

4.2 3DES XOR P1

3DES or the Triple Data Encryption Algorithm (TDEA) was
32 bits F
developed by IBM in 1998-1999 to address the obvious flaws
8 bits 8 bits 8 bits 8 bits
in DES without designing a whole new cryptosystem. The 56-
bit key used in DES is not deemed secure enough to encrypt S1 S2 S3 S4
sensitive data. 3DES simply extends the key size of DES by 32 bits 32 bits 32 bits 32 bits
applying the algorithm three times in succession with three
different keys. The combined key size is thus 168 bits. TDEA 32 bits XOR
involves using three 64-bit keys (K1, K2, K3) in Encrypt- 32 bits
Decrypt- Encrypt (EDE) fashion, that is, the plain text is
32 bits
encrypted with K1, then decrypted with K2, and then XOR
encrypted again with K3 [5]. The encryption process can be
expressed as C = E(D(E(P, K1), K2), K3), similarly,
decryption process can be interpreted as P = D(E(D(C, K3),
K2), K1) Li Ri
The standards define three keying options:
 Option 1: the strongest and preferred option,
employs three mutually independent key, with 3 ×
56 = 168 independent key bits.
L16 R16
 Option 2: K1 and K2 are independent, and K3 = K1,
this provides a shorter key length of 2 × 56 = 112
bits and a reasonable compromise between DES and XOR P16
Keying option 1, but NIST has deprecated this F
option.
 Option 3: All three keys are identical. This option is S1 S2 S3 S4
equivalent to DES Algorithm and NIST no longer
allows this option.
XOR
While the 3-times iteration is applied to increase security, it is
tripled the processing time as well, making 3DES one of the
slowest block cipher methods. XOR

4.3 BLOWFISH
Blowfish was designed in 1993 by Bruce Schneier as a fast, XOR P18 P17 XOR
free patent alternative to existing encryption algorithms at that
time. Blowfish provides a good encryption rate and no L17 R17
cryptanalysis of it has been recorded to date. The algorithm
consists of two parts: a key-expansion part and a data-
encryption part. Blowfish uses a large number of subkeys Cipher Text
which are constructed during the Key expansion phase. This
phase converts a variable-length key of at most 56 bytes (448 Figure 4: Blowfish Algorithm
bits) into an array of sub-keys called P-array consists of Blowfish algorithm is as follows:
eighteen 32-bit sub-keys totaling 4168 bytes. There are also
four 32-bit S-boxes. The keys must be computed before any 1. The 64-bit plain text is split into two halves L0 and
data encryption or decryption [6]. The creation of sub-keys R0.
further increases security, because a hacker would have to 2. For each round i = 1, 2, …, 16
crack more than just the original key.
a. The left half L is XORed with the ith sub key
The second part of the Blowfish routine that is a data L i = L i ⊕ Pi
encryption is done through 16 Fiestel network rounds, a swap
operation and two exclusive-or operations. The F function b. The XOR of F(Ri, Ki) and the left half L is
takes the 32-bit input and divides it into 4 bytes (8-bits each). computed yielding Ri = F(Li)⊕ Ri
These four values are then used for table lookup in their
corresponding S-Boxes. A graphic representation of Blowfish c. The two sides are swapped to set the R and L
is given in Figure 4. for the next round, that is Li+1 = Ri and Ri+1 =
Li ⊕ F(Ri, Ki)
3. Swapp R16 and L16 again to undo the last swap.
4. R17 = R17 ⊕ P17, L17 = L17 ⊕ P18
5. L and R are recombined to construct the cipher text.

9
 International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 48, April 2019

The F function looks like this: Divide L into four eight-bit bit.
quarters: a, b, c, and d. Then, F(L) = (((((S + S2) mod 232)
XOR S3) + S4) mod 232). One MASHING round is defined as follows, for each i = 0, 1,
2, and 3:
The complexity of the Blowfish key generating process made
it a considerable robust block cipher algorithm [7]. R[i] = R[i] + K[ R[i − 1] & 63 ];
The entire encryption operation with RC2 can now be
4.4 TWOFISH described as follows [8]:
Twofish is a symmetric block cipher with block size of 128
bits, and accepts a key of any length up to 256 bits. Twofish is 1. Initialize words R[0], ..., R[3] to contain the 64-bit
fast on both 32-bit and 8-bit CPUs. It can be used in network plaintext block.
applications where keys are changed frequently and where 2. Expand the key, so that words K[0], ..., K[63]
little or no RAM and ROM available. Twofish is a Feistel become defined.
network [8]. This means that in each round, half of the text
block is sent through an F function, and then XORed with the 3. Initialize j to zero.
other half of the text block.
4. Perform five MIXING rounds.
In each round of Twofish, two 32-bit words serve as input into
the F function. Each word is broken up into four bytes. Those 5. Perform one MASHING round.
four bytes are sent through four different key-dependent S- 6. Perform six MIXING rounds.
boxes. The four output bytes (the S-boxes have 8-bit input and
output) are combined using a Maximum Distance Separable 7. Perform one MASHING round.
(MDS) matrix and combined into a 32-bit word. Then the two 8. Perform five MIXING rounds.
32-bit words are combined using a Pseudo-Hadamard
Transform (PHT), added to two round subkeys, then XORed 9. The ciphertext is R[0], ..., R[3].
with the right half of the text. There are also two 1-bit
Although no design weaknesses have been identified in RC2
rotations going on, one before and one after the XOR.
which could lead to practical cryptanalytic attacks, it is
Twofish also has something called "Prewhitening" and
considered to be a slow cipher.
"Postwhitening;" additional subkeys are XORed into the text
block both before the first round and after the last round. 4.6 RC5
RC5 is a 32/64/128-bit block cipher developed in 1994. It was
4.5 RC2 designed also by Ronald Rivest in December of 1994. It is a
RC2 is a 64-bit block cipher with variable key size designed
symmetric block cipher having a variable number of rounds,
by Ron Rivest in 1987. "RC" stands for "Ron's Code" or
word size and a secret key.
"Rivest Cipher". The plaintext is divided into four words each
of 16 bits. There is a key expansion process before the As in RC2 there is a key-expansion routine that expands the
encryption or decryption take place. The key expansion user's key K to fill an expanded key array S. It is assumed that
process takes a user key between 1 and 128 bytes in length the input block is given in two w-bit registers A and B, and
together with a parameter that specifies the effective key- that key-expansion has already been performed, so that the
length of encryption, then an array K of 64 sub keys is derived array S[0...t - 1] has been computed. Here is the encryption
(16-bit each). Then encryption is done through two kinds of algorithm in pseudo-code [9]:
rounds. One is termed a MIXING round and the other a
MASHING round. There are in total 16 mixing rounds and 1. A = A + S[0];
two mashing rounds. In each round each of the four words in 2. B = B + S[1];
the intermediate cipher texts is updated as a function of the
other words. Each of the mixing rounds takes a 16-bit subkey. 3. For i = l to r do

An array of four 16-bit words R[0], ..., R[3] is used to hold the a. A = ((A | B) <<< B) + S[2 * i];
initial plaintext, the intermediate results, and the final cipher b. B = ((B | A) <<< A) + S[2 * i +1];
text.
The output is in the registers A and B.
One MIXING round is defined as follows, for each i = 0, 1, 2,
and 3: It is a simple algorithm which has a low memory requirement.
It is fast and yet secure if suitable parameters are chosen.
1. R[i] = R[i] + K[j] + (R[i − 1] & R[i − 2]) + (∼R[i −
1] & R[i − 3]); 4.7 RC6
Where & denotes bitwise logical and, ⊕ denotes bitwise It was one of the AES finalist developed also by Ronald
XOR and ∼ denotes bitwise complementation. All 16-bit Rivest in 1997. It uses 128 bit block size and supports key
word addition + is performed modulo 216. sizes of 128, 192 and 256 bits. It is an improvement of the
RC5 Algorithm and provides even better security. It makes
2. j = j + 1; use of 4 registers (Each one of 32 bit). It uses fewer rounds
and offers a higher throughput. The following is the
Here j is a “global” variable so that K[j] is always the first key
encryption algorithm in pseudo-code, where S is the array of
word in the expanded key which has not yet been used in a
keys generated by the key expansion routine:
MIXING operation.
3. R[i] = R[i] <<< s[i];
where s[0] = 1, s[1] = 2, s[2] = 3, and s[3] = 5. And
R[i] <<< s[i] denote that R[i] is rotated left by s[i]

10
 International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 48, April 2019

1. B = B + S[0]; Column of the states. A fix matrix is multiplied to

each column vector.
2. D = D + S[1];
4) Addroundkey transformation: It is a bitwise XOR
3. for i = 1 to r do between the 128 bits of current state and 128 bits of
a. t = (B*(2B + 1)) <<< lg w the round key. This transformation is its own
inverse
b. u = (D*(2D + 1)) <<< lg w
According to many literature reviews [14, 15, 16, 17, 18],
c. A = ((A ⊕ t) <<< u) + S[2i] AES is considered one of the most secure algorithms. In
d. C = ((C ⊕ u) <<< t) + S[2i + 1] theory it has not broken since the time of its development.

e. (A, B, C, D) = (B, C, D, A) 5. COMPUTATIONAL RESULTS

The main goal of any encryption algorithm is security.
4. A = A + S[2r + 2]; However, other performance metrics may determine the
5. C = C + S[2r + 3]; preference use of one algorithm over another. In this section,
we compare the algorithms on the basis of different
4.8 AES parameters, the structure of the algorithm, encryption and
The Advanced Encryption Standard (AES), also known by its decryption times, throughput and memory usage. The
original name Rijndael, is the encryption standard experimental results are depicted against varied file sizes and
recommended by NIST in 2001. AES is developed by tow recording the computation cost for those algorithms.
Belgain cryptographers, Vincent Rijmen and Joan Daemen,
who submitted a proposal to NIST during the AES selection 5.1 The Structure of the Algorithm
process. Unlike the other cipher algorithms discussed in this The structure is characterized by a set of variables such as the
paper, AES does not use a Feistel network. Instead AES is a block size, key size, and number of rounds. Ultimately, these
variant of Rijndael ciphers which has a fixed block size of 128 are the factors which affect the security of a particular
bits, and a key size of 128, 192, or 256 bits. Rijndael is algorithm. The key size plays a vital role in algorithm
specified with block and key sizes that may be any multiple of security. Larger key size provides higher security when other
32 bits, with a minimum of 128 and a maximum of 256 bits. factors were considered to be equal in some algorithms.
AES algorithm can support any combination of data (128 bits) Another critical factor is the number of rounds used for
and key length of 128, 192, and 256 bits. The key size used encryption/decryption process. Performing more rounds,
for an AES specifies the number of rounds needed to encrypt strengthens the security of the algorithm, but increases the
or decrypt the text. AES goes through 10 rounds for 128-bit complexity as well. That is why, when designing a
keys, 12 rounds for 192-bit keys, and 14 rounds for 256-bit cryptographic algorithm, the number of rounds is one
keys in order to deliver final cipher text or to retrieve the important factor that should be set carefully. Table 1
original plain text. AES allows a 128 bit data length that can summarizes this comparison.
be divided into four basic operational blocks. These blocks are
treated as array of bytes and organized as a matrix of the order 5.2 B. Encryption and Decryption Tim
of 4×4 that is called the state. For both encryption and The time needed by every algorithm to encrypt and decrypt
decryption, the cipher begins with an AddRoundKey stage. varied size files is calculated and analyzed. Table 2 and Table
However, before reaching the final round, this output goes 3 are graphically represented by Figure 5 and Figure 6, they
through nine main rounds, during each of those rounds four give the results of time in milliseconds of encryption and
transformations are performed; sub-bytes, shift-rows, mix- decryption processes, respectively. Encryption and decryption
columns, and add round Key. In the final (10th) round, there rates are dependent on the processor speed, and algorithm
is no mix-column transformation [2]. Decryption is the structure. From the tabular results, it is easy to observe that
reverse process of encryption and using inverse functions: AES, RC6 and RC2 have the best encryption and decryption
Inverse Substitute Bytes, Inverse Shift Rows and Inverse Mix performance over other algorithms in terms of time
Columns. consumption. On the other hand, the complexity of 3DES is
clearly reflected on the time it takes to encrypt and decrypt
The date block goes through four steps of transformation at files.
each round [10, 11, 12, 13]:
5.3 Throughput
1) Substitute Byte transformation: AES contains 128
The throughput of an algorithm is calculated by dividing the
bit data block, which means that each of the data
total block size in Megabyte encrypted or decrypted on the
blocks has 16 bytes. In sub-byte transformation,
total encryption or decryption time. The throughput of the
each byte (8-bit) of a data block is transformed into
encryption scheme is calculated using the following formula:
another block using an 8-bit substitution box which
is known as Rijndael S-box. Throughput = Total Plain Text / Encryption Time
2) Shift Rows transformation: It is simple byte Similarly, the decryption scheme is calculated using a similar
transpositions, the bytes in the last three rows of the formula:
state, depending on the row location, are cyclically
shifted. For 2nd row, 1 byte circular left shift is Throughput = Total Plain Text / Decryption Time
performed. For the 3rd and 4th row 2-byte and 3- The throughput of encryption and decryption processes is
byte left circular left shifts are performed given in Figure 7 and Figure 8, respectively. The figures show
respectively. the superiority of Twofish and Blowfish over the other
3) Mixcolumns transformation: This round is algorithms.
equivalent to a matrix multiplication of each

11
 International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 48, April 2019

5.4 Memory Utilization Blowfish, Twofish, RC2, RC5, RC6, and AES. Each
Memory utilization or memory usage is another important algorithm aims to introduce extra level of security and to
parameter that should be taken into consideration. The cost of satisfy performance requirements more than earlier proposed
memory often exceeds the cost of CPU. As a result the algorithms solution. This extra security level should balance
efficient utilization of memory has received much attention between a robust algorithm structure and a reasonable
when designing an algorithm. The memory requirement complexity computation. This work provides a detailed
depends on the key size, initialization vectors, and type of review of the structure of the mentioned algorithms along
operations. Table 4 and Table 5 give memory utilization for with a performance evaluation comparison that highlights the
encryption and decryption processes for each algorithm, strengths and limitations of each algorithm. From the tabular
respectively, while Figure 9 and Figure 10 graphically comparison, some key points can be concluded. First, the
interprets such values. Twofish and Blowfish algorithms outperform the others
followed by RC6 and AES, while 3DES and RC2 come at the
6. CONCLUSION tail of the list. Second, when it comes to memory usage, there
Data security and privacy have been always the main concern is no significant difference between most of the algorithms
about cloud computing services. Cryptography is a main trend except for 3DES which swept over almost three times of the
to achieve data security and privacy. Since the introduction of memory needed for the other algorithms. Third, while AES is
cloud computing technology, several cryptography solutions considered one of the most secure algorithms and no attacks
have been proposed for protecting outsourced data and user have been reported against it, Twofish and Blowfish are
privacy and for insuring that data are not being leaked to considered the fastest schemes for both encryption and
unauthorized third party. Most of these solution schemes aim decryption.The main goal of any encryption algorithm is
at achieving a tradeoff between security and functionality. security. However, other performance metrics may determine
This paper represents a comprehensive study of the major the preference use of one algorithm over another. In this
symmetric key block cipher algorithms, namely, DES, 3DES, section,
Table 1. Comparision of Stmmetric Algothims Structure

Algorith Structure metrics

m Block size Key size Rounds
DES 64 bits 56 bits (+8 parity bits) 16
3DES 64 bits 168, 112 or 56 bits 48 DES-equivalent rounds
Blowfish 64 bits 32–448 bits 16
Towfish 128 bits 128, 192 or 256 bits 16
16 of type MIXING, 2 of type
RC2 64 bite 8–1024 bits; default 64 bits
MASHING
RC5 32, 64 or 128 bits 0 to 2040 bits 1-255
RC6 128 bits 128, 192, or 256 bits 20
AES 128 bits 128, 192 or 256 bits 10, 12 or 14 (depending on key size)

Table 2. Encryption Time

Encryption Time in milliseconds
File Type File Size
DES 3DES Blowfish Twofish RC2 RC5 RC6 AES
JPG 96 KB 569.32 626.58 511.82 511.39 634.66 582.12 516.61 552.67
TXT 116 KB 737.1 698.61 654.13 646.12 719.81 759.39 666.56 756.4
PDF 324 KB 1792.7 1930.89 1655.37 1646.31 1938.56 1818.06 1698.27 1735.09
PPT 1.29 MB 6658.48 8121.04 6513.87 6480.82 8131.14 6646.65 6539.34 6569.1
MP3 2.04 MB 10762.89 11589.72 10293.91 10279.16 11886.39 10561.62 10331.95 10612.97
MP4 2.13 MB 11189.7 12209.9 10610.74 10600.05 12503.65 11057.51 10797.49 10820.48
5285.031
Average Time 5862.79 5039.97 5027.308 5969.035 5237.558 5091.703 5174.451
6
Throughput in MB/s 0.18869 0.17009 0.19786 0.198364 0.16706 0.19040 0.19585 0.19272

12
 International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 48, April 2019

Table 3. Decryption Time

Decryption Time in milliseconds
File Type File Size
DES 3DES Blowfish Twofish RC2 RC5 RC6 AES
JPG 96 KB 546.52 601.45 488.45 485.12 607.75 493.01 487.47 492.23
TXT 116 KB 653.26 669.18 599.07 588.11 679.67 610.97 609.05 615.84
PDF 324 KB 1803.27 1938.56 1659.71 1617.94 2178.89 1682.85 1645.62 1660.3
PPT 1.29 MB 6740.09 6742.68 6519.72 6408.95 6807.16 6793.51 6589.45 6667.7
MP3 2.04 MB 10406.47 10830.28 10331.07 10330.41 10899.95 10493.83 10431.84 10449.53
MP4 2.13 MB 11211.71 11401.17 10741.26 10699.38 11405.26 10856.3 10867.01 10966.1
Average Time 5226.88 5363.88 5056.54 5021.65 5429.78 5155.07 5105.07 5141.95
Throughput in MB/s 0.19079 0.18591 0.19721 0.19858 0.18366 0.19344 0.19534 0.19394

Figure 5: Encryption Time

Figure 6: Decryption Time

13
 International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 48, April 2019

Figure 7: Throughput of encryption algorithm

Figure 8: Throughput of decryption algorithms

Table 4. Memory Utilization for Encryption Algorithm
Memory Utilization for Encryption in Kilobytes
File Type File Size
DES 3DES Blowfish Towfish RC2 RC5 RC6 AES
JPG 96 KB 133.72 883.59 134.36 134.36 232.58 133.02 133.78 134.34
TXT 116 KB 224.62 883.59 225.7 225.7 324.84 225.59 224.73 225.68
PDF 324 KB 450.71 1343.13 452.85 452.85 551.13 340.13 450.91 340.29
PPT 1.29 MB 1456.36 3534.21 1463.37 1463.37 1957.78 1462.66 1457.07 1463.37
MP3 2.04 MB 2240.52 4417.75 2251.3 2251.3 2842.71 2250.21 2241.62 2251.3
MP4 2.13 MB 2342.27 5301.28 2353.6 2353.6 2944.61 2352.34 2343.54 2353.59

14
 International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 48, April 2019

Table 5. Memory Utilization for Decryption Algorithm

Memory Utilization for Decryption in Kilobytes
File Type File Size
DES 3DES Blowfish Towfish RC2 RC5 RC6 AES
JPG 96 KB 123.53 883.59 124.13 126.16 142.38 122.79 122.32 124.13
TXT 116 KB 224.62 883.59 225.7 225.7 324.84 225.59 224.73 225.66
PDF 324 KB 448.15 1767.13 450.3 450.29 548.585 337.58 448.37 450.3
PPT 1.29 MB 1456.36 4417.75 1463.37 1463.37 1757.78 1462.66 1457.07 1463.37
MP3 2.04 MB 2189.59 7068.36 2200.21 2200.21 2891.77 2198.96 2190.68 2200.21
MP4 2.13 MB 2316.9 7068.36 2328.07 2328.05 2919.13 2326.85 2318.12 2328.07

Figure 9: Memory Utilization for encryption algorithms

Figure 10: Memory Utilization for decryption algorithms

7. REFERENCES [4] privacy." Scientific American, vol. 228, no. 5, pp. 15-2,
[1] NIST SP 800-145, “A NIST definition of cloud 1973.
computing”, [5] FIPS PUB 46-3, “Data encryption standared (DES),”
https://csrc.nist.gov/publications/detail/sp/800-145/final, National Bureau of Standards, U.S. Department of
September 2011. Commerce, January 1977.
[2] W. Stalling, “Cryptography and network security [6] D. Coppersmith · D. B. Johnson · S. M. Matyas, “A
principles and practices,” Fourth Edition, Prentice Hall, proposed mode for triple-DES encryption,” IBM Journal
December 2006. of Research and Development, April 1996.
[3] H. Feistel, "Cryptography and computer

15
 International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 48, April 2019

[7] B. Schneier, "The Blowfish Encryption Algorithm," Dr. (RSA, DES, 3DES and AES) for information security,”
Dobb's Journal , v.19, n. 4, pp. 38-40, April 1994. International Journal of Computer Applications, vol. 67,
no. 19, pp. 33-38, April 2013.
[8] B. Schneier, "Description of a New Variable-Length
Key, 64-bit Block Cipher (Blowfish)," Fast Software [14] M. Faheem Mushtaq, S. Jamel, A. Disina, Z. Pindar, N.
Encryption: Second International Workshop, Leuven, Shafinaz A. Shakir, M. Deris, " A Survey on the
Belgium, Proceedings, Springer-Verlag, pp.191-204, Cryptographic Encryption Algorithms", International
December 1994. Journal of Advanced Computer Science and
Applications, vol. 8, no. 11, 2017.
[9] L. Knudsen, V. Rijmen, R. Rivest, and M. Robshaw, "On
the Design and Security of RC2", Fast Software [15] J. Daemen and V. Rijmen, "AES Proposal: Rijndael,"
Encryption, 5th International Workshop, FSE '98, Paris, AES algorithm submission, September 1999.
France, March 23-25, 1998.
[16] M. Ebrahim, S. Khan, U. Khalid, “Symmetric algorthm
[10] R. Rivest, " The RC5 encryption algorithm", survey: A comparative analysis”. International Journal of
International Workshop on Fast Software Encryption, pp Computer Applications, vol. 61, no. 20, pp. 12-19,
86-96, 1994. January 2013.
[11] B. Schneier, J. Kelsey, D. Whiting, D. Wagner, C. Hall, [17] B. Schneier and D. Whiting, “A performance comparison
and N. Ferguson. “Twofish: A 128-bit block cipher.” In of the five AES finalists.” Proceedings of the Third AES
AES Round 1 Technical Evaluation CD-1: Candidate Conference, pp. 123-135, April 2000.
Documentation. National Institute of Standards and
Technology NIST, August 1998. See http://www. [18] Z. Hercigonja and D. gimnazija, “Comparative analysis
nist.gov/aes of cryptographic algorithms.” International Journal of
Digital Technology & Economy, vol.1, no. 2, 2016.
[12] A. Mandal, C, Parakash and A. Tiwari, “Performance
evaluation of cryptographic algorithms: DES and AES”, [19] D. Abdul Elminaam et al., "Performance Evaluation of
IEEE Students’ Conference on Electrical, Electronics Symmetric Encryption Algorithms", International
and Computer Science, pp. 1-5, 2012. Journal of Computer Science and Network Security,
vol.8 no.12, December 2008.
[13] G. Singh and Supriya, “A study of encryption algorithms

IJCATM : www.ijcaonline.org 16