Hash Computation Research Articles

Secure and Efficient Authentication using Linkage for permissionless Bitcoin network

The cryptocurrency’s permissionless and large-scale broadcasting requirements prohibit the traditional authentication implementation on the blockchain’s underlying peer-to-peer (P2P) networking. Thus, blockchain networking implementations remain vulnerable to networking integrity threats such as spoofing or hijacking. We design Secure and Efficient Authentication using Linkage (SEAL) to build connection security for permissionless Bitcoin networking. SEAL uses the linkage between the packets for a symmetric operation, in contrast to the traditional authentication approach relying on identity-credential-based trust. To make it appropriate for cryptocurrency networking, SEAL utilizes the packet header, protects the end-to-end connection, and separates the online process and the offline process so that the real-time overhead is minimal for greater efficiency and practicality. We implement SEAL on a functioning Bitcoin node and demonstrate that SEAL operates efficiently with minimal overhead. Specifically, it reduces the hash rate by only 1.3% compared to an unsecured node. Additionally, we use a network simulator to emulate the Bitcoin Mainnet and analyze SEAL’s impact on block propagation delay. SEAL yields 2.04 times smaller delay and 1.25 times smaller delay in block propagation than HMAC and ChaCha20-Poly1305, respectively. The key advantage of SEAL is that it requires fewer hash computations and simpler mixing operations, resulting in significantly lower computational overhead compared to traditional authentication schemes based on message authentication codes (MACs).

A Low-Complexity Security Scheme for Drone Communication Based on PUF and LDPC

Due to the limited payload and power of drones, the computational overhead, storage overhead and communication overhead that can be used for secure communication are restricted, making it difficult to apply some complex but fairly secure authentication protocols on drones. In this paper, we propose a low-complexity protocol for storing identity information in a resource-unconstrained device that does not require the UAV to store the information, thereby enhancing the UAV’s resistance to capture. The protocol in this paper mainly consists of quasi-cyclic low-density parity-check (QC-LDPC) codes, physical unclonable functions (PUFs) based on random-access memory (RAM), “XOR” operations, and hash computation. The protocol in this paper is an authentication architecture in which the drone is guided by the ground station to read its identity information, and the drone does not store any identity information in advance. The protocol is divided into two phases: 1. fuzzy authentication of fingerprint PUF and 2. uniqueness authentication accomplished while guiding the recovery of identity PUF. Recovering identity PUF in this paper, QC-LDPC is used as the error control module, and the optimization of bit-flip decoding significantly reduces the probability of decoding failure. After the comparative security analysis and comparative overhead analysis of this paper’s protocol, it can be concluded that this paper’s protocol can withstand common attacks (including attacks attempting to pass authentication, attacks attempting to interfere with authentication, and physical capture attacks), and the storage and communication overhead is small in the case of large time overhead.

SHA-256 Hash Function on Intel DE10 Lite FPGA

Abstract: The SHA-256 hash function is a standardized and trusted algorithm that takes a set of data and produces a unique, deterministic, and irreversible representation called a hash or digest. A component of other protocols, SHA-256 protects password storage, secures and verifies Bitcoin transactions, and authenticates internet communication. We did a thorough analysis of the hash function and a preexisting Verilog implementation at the algorithmic, architectural, and circuit levels to identify and address the bottlenecks. We propose a new SHA-256 hardware architecture that utilizes binary tree structured adder trees to speed up hash computation. The proposed design targets Intel DE10-Lite FPGA and achieves 23% increase in computation speed. In applications, this can offer faster online communication or a more secure Bitcoin network.

High-Performance Encryption Algorithms for Dynamic Images Transmission

With the proliferation of the internet, the issue of image tampering has escalated, necessitating robust image encryption schemes. Despite the multitude of proposed image encryption algorithms, challenges such as slow computational speed, weak security, and suboptimal visibility persist. This study addresses these challenges by introducing a high-performance encryption algorithm tailored for dynamic images—QEDI (Quick Encryption Algorithm for Dynamic Images). QEDI leverages elliptic curve-based asymmetric encryption algorithms and hash algorithms. The process involves using a hash algorithm to calculate the hash of the dynamic image slated for encryption, employing the elliptic curve algorithm to generate public and private keys, and utilizing the asymmetric encryption algorithm to process the image hash, resulting in a signature ciphertext. This ciphertext is embedded into the dynamic image, completing the encryption process. To enhance the execution speed without compromising the image quality and visibility, the bit depth of the dynamic image is reduced before hash calculation, and custom information fields are encapsulated to embed the signature ciphertext into the dynamic image. Experimental evaluations, conducted within a dynamic image generation system, encompassed assessments of security, encryption algorithm execution time, and image quality. The results indicate that utilizing QEDI for dynamic image encryption yields better security compared to EDI (Encryption Algorithm for Dynamic Images). QEDI exhibits minimal impact on image quality, with a noteworthy 69.77% reduction in execution time compared to EDI. Furthermore, when compared to existing algorithms employing image conversion for encryption, QEDI-encrypted images demonstrate enhanced visibility, better image quality, and expedited encryption speed.

Blockchain-Secure Mobility Data in Smart Campus

Acquiring knowledge of the patterns of human mobility within a university setting is a critical endeavor that can facilitate the development of effective strategies for future work programs. It is essential to ascertain the positions of campus inhabitants as they engage in daily activities that align with the institution’s work plan framework. Nonetheless, the paramount challenge associated with the presence of personal data on human movement is ensuring the utmost security of this sensitive information. Blockchain technology offers a solution by enabling the safeguarding of personal data through the decentralization of information, wherein individuals act as controllers in a distributed cloud network. In the present study, a straightforward system comprising GPS sensors and a Raspberry Pi is employed to detect personnel’s location data. The SHA256 algorithm is utilized to generate a hash that connects the constituent blocks, thereby significantly enhancing data security. The intricate hash computation is validated through the implementation of proof-of-work, which generates pertinent binary data at an expeditious block mining time of 16.64 milliseconds. This approach effectively thwarts cyberattacks and ensures the maximum protection of data.

Distributed and Lightweight Software Assurance in Cellular Broadcasting Handshake and Connection Establishment

With developments in OpenRAN and software-defined radio (SDR), the mobile networking implementations for radio and security control are becoming increasingly software-based. We design and build a lightweight and distributed software assurance scheme, which ensures that a wireless user holds the correct software (version/code) for their wireless networking implementations. Our scheme is distributed (to support the distributed and ad hoc networking that does not utilize the networking-backend infrastructure), lightweight (to support the resource-constrained device operations), modular (to support compatibility with the existing mobile networking protocols), and supports broadcasting (as mobile and wireless networking has broadcasting applications). Our scheme is distinct from the remote code attestation in trusted computing, which requires hardwarebased security and real-time challenge-and-response communications with a centralized trusted server, thus making its deployment prohibitive in the distributed and broadcasting-based mobile networking environments. We design our scheme to be prover-specific and incorporate the Merkle tree for the verification efficiency to make it appropriate for a wireless-broadcasting medium with multiple receivers. In addition to the theoretical design and analysis, we implement our scheme to assure srsRAN (a popular open-source software for cellular technology, including 4G and 5G) and provide a concrete implementation and application instance to highlight our scheme’s modularity, backward compatibility to the existing 4G/5G standardized protocol, and broadcasting support. Our scheme implementation incorporates delivering the proof in the srsRAN-implemented 4G/5G cellular handshake and connection establishment in radio resource control (RRC). We conduct experiments using SDR and various processors to demonstrate the lightweight design and its appropriateness for wireless networking applications. Our results show that the number of hash computations for the proof verification grows logarithmically with the number of software code files being assured and that the verification takes three orders of magnitude less time than the proof generation, while the proof generation overhead itself is negligible compared to the software update period.

FIMAR: Fast incremental memory acquisition and restoration system for temporal-dimension forensic analysis

Recent advances in anti-forensic techniques, including encryption, obfuscation, and kernel-structure manipulations, make conventional storage, network, and memory forensics difficult. For example, the code of packed malware is encrypted on storage, network packets, and system RAM to hide its existence; it appears briefly in plain text on the system RAM during its execution phase. Memory forensics, therefore, has been a vital component of today's digital forensics. Conventional memory acquisition techniques, however, capture only the current state of a system RAM; if the timing of the acquisition is late, the evidence could be lost forever. Since the state-of-the-art anti-forensic attacks leave only the slightest evidence for the shortest time, finer temporal acquisition and analysis are needed. This paper presented a Fast Incremental Memory Acquisition and Restoration (FIMAR) system that enables us to acquire multiple memory snapshots (i.e., memory dumps) over time and to construct a timeline of past activities on a computer. A thin hypervisor and its Second Level Address Translation (SLAT) were used to track memory page changes and to obtain atomic memory snapshots. The hash calculation of 4 KiB chunks was used to detect updated parts of the system RAM to reduce the transfer size. We implemented the incremental memory acquisition algorithms using a thin hypervisor called BitVisor. Finally, we tested the developed FIMAR system to analyze BlueSky ransomware with anti-forensic logic. The developed FIMAR system improved the atomicity of memory snapshots using TLB shootdown and considering per-core Extended Page Tables (EPTs) compared to previous hypervisor-based memory acquisition systems.

LadderFilter: Filtering Infrequent Items with Small Memory and Time Overhead

Data stream processing is critical in streaming databases. Existing works pay a lot of attention to frequent items. To improve the accuracy for frequent items, existing solutions focus on accurately filtering infrequent items. While these solutions are effective, they keep track of all infrequent items and require multiple hash computations and memory accesses. This increases memory and time overhead. To reduce this overhead, we propose LadderFilter, which candiscard infrequent items efficiently in terms of both memory and time. To achieve memory efficiency, LadderFilter discards (approximately) infrequent items using multiple LRU queues. To achieve time efficiency, we leverage SIMD instructions to implement LRU policy without timestamps. We apply LadderFilter to four types of sketches. Our experimental results show that LadderFilter improves the accuracy by up to 60.6×, and the throughput by up to 1.37×, and can maintain high accuracy with small memory usage. All related code is provided open-source at Github.

SimBlock Simulator Enhancement with Difficulty Level Algorithm Based on Proof-of-Work Consensus for Lightweight Blockchain

Proof-of-Work (PoW) was the first blockchain consensus protocol discovered, followed by Proof-of-Stake and others. The disadvantage of the PoW is that it requires high energy consumption compared to other consensus protocols. Based on this weakness, some researchers proposed a lightweight blockchain technology, a modified blockchain that has a simplified algorithm but does not reduce the security factor. This lightweight blockchain is suitable for applications requiring data reliability but with limited computing resources, such as Internet of Things devices. This paper discussed and modified the SimBlock simulator as one of the existing blockchain simulators. It has a visualization tool to look further into the propagation transition of the block. Unfortunately, the existing PoW consensus on the SimBlock simulator is unable to pinpoint the actual hash computation method. Therefore, the hashing process in the SimBlock simulator was modified by including the difficulty level for finding the hash target. The purpose of including the difficulty level was to determine how long it takes to create a block. By knowing the time needed, a recommendation could be obtained for the most suitable difficulty level for a lightweight blockchain and its implementation with IoT devices. There are two options of approaches to the difficulty level referred to in this paper; finding the number of zeros that appear sequentially and are in front of a hash value (leading zero) and finding the number of zeros that appear arbitrarily (count zero). For example, the first difficulty level on a leading-zero quest has the same meaning as searching for a leading zero, the second level of difficulty is the search for the two leading zeros, etc. The block generation time on a blockchain network using the PoW consensus highly depends on the difficulty level. Block generation time and resource utility have been analyzed and compared with other blockchain simulators and existing networks, such as Ethereum and Bitcoin. The modified SimBlock simulator was tested in this experiment using 100-600 nodes, with the expected result of creating 100-1000 blocks. Based on the experiments, creating a block using leading zeros as the hash target for the first to fourth difficulty levels took less than 1 s, whereas when using count zeros (zero-count) as the target hash, it took less than 1 s for the first to fifteenth difficulty levels. Using leading zeros took approximately 237.4 s at difficulty level 7, while count-zero took approximately 633.8 s at difficulty level 19. The experiment was not continued at the next difficulty level because it required a longer compilation time. With the modifications made, the creation of a block on a blockchain network using the PoW consensus can be clearly seen. The difficulty level added to the simulator can also provide information for determining the difficulty level to be implemented on the lightweight blockchain.

HB+-MHT: Lightweight and Efficient Data Integrity Verification Scheme for Cloud Virtual Machines

With the rapid development of cloud computing, cloud storage is widely used. In the cloud environment, users’ virtual machine system mirrors and data are stored in the cloud server. The escape of virtual machines and Trojan virus attacks make it challenging to ensure the integrity of virtual machine systems. Trusted computing is expensive to randomly verify data integrity and does not adapt to dynamic data changes. Provable data integrity is a potential solution to this problem. Merkle Hash Tree (MHT) model is widely adopted in provable data integrity. Although MHT requires only a small amount of evidence for verification, the verifier’s number of hash calculations and the server’s efficiency of evidence query are not optimal. Moreover, the verification frequency of each piece of data is not considered by MHT. Properly handling these factors can improve the actual verification performance. In this paper, a lightweight and efficient data integrity verification approach called HB+-MHT is proposed for the tenant virtual machine (TVM) in cloud computing. In HB+-MHT, the Huffman hash tree scheme is used for small file verification to ensure that the hot file has a shorter path, which reduces the required amount of evidence for verification. Meanwhile, the B+ hash tree scheme is used for big files verification, which can effectively reduce evidence query time and hash calculation times. The experimental results show that the scheme proposed in this paper can perform data integrity verification well, with reduced computing and storage overhead.

Robust Image Hashing With Saliency Map And Sparse Model

Abstract Image hashing is an effective technology for extensive image applications, such as retrieval, authentication and copy detection. This paper designs a new image hashing scheme based on saliency map and sparse model. The major contributions are twofold. The first contribution is the construction of a weighted image representation by combining a visual attention model called Itti model and the matrix of color vector angle (CVA). Since the Itti model can efficiently detect saliency map and CVA fully captures color information of image, they contribute to a visually robust and discriminative image representation. The second contribution is the hash extraction from the weighted image representation via sparse model. A classical sparse model called robust principal component analysis is exploited to decompose the weighted image representation into a low-rank component and a sparse component. As the low-rank component can describe intrinsic structure of image, hash calculation with low-rank component can achieve good discrimination. The efficiencies of the proposed scheme are validated by extensive experiments with open databases. The results demonstrate that the proposed scheme is superior to some state-of-the-art schemes in terms of classification performance between robustness and discrimination.

A blockchain and smart contract-based data provenance collection and storing in cloud environment

Data uploading needs security and privacy in the cloud. But there are some problems like centralized provenance data (PD) collection, storage, lack of security, integrity, and more time consumption. There are methods like Rabin, Knapsack, McEliece, Elagamal, and Rivest–Shamir–Adleman for the generation of keys but it increases the encryption and decryption time and less security. Therefore, the blockchain and smart contract-based data provenance (BSCDP) Architecture is proposed for providing secure storage in the cloud environment. Initially, the fingerprint biometrics and physically uncloneable functions (PUF) have been used in verification process. The combination of PUF and fingerprint biometrics is used for secure data transmission. To protect privacy and strengthen the security, the fuzzy extractor is employed. Secondly, we use an elliptic-curve key based cyclic shift transposition cryptography algorithm for enhancing the security when sharing the key. Thirdly, we introduce the blockchain and the interplanetary file system (IPFS) for PD collection, hash computation, and storing with reduced computational overhead (CO). The integrity of data is maintained by using blockchain based secure hashing algorithm-3. By arranging fuzzy based smart contracts (FSC), the data user (DU) tracks their data. FSC is employed for tracking the history of data. The data collected is directly stored in IPFS and the DU gets a hash from IPFS to retrieve the data in the future. Finally, the data verification is done by the provenance auditor. When comparing our proposed BSCDP method with existing methods, the proposed BSCDP method achieves high security in the cloud environment for 2 K users in terms of evidence insertion time (30 ms), verification time (30 ms), response time (40 ms), total change rate (5%), CO (5.8 Kb), encryption (50 ms) and decryption time (52 ms).

Energy-efficient distributed password hash computation on heterogeneous embedded system

This paper presents the improved version of our cool Cracker cluster (cCc), a heterogeneous distributed system for parallel and energy-efficient bcrypt password hash computation. The cluster consists of up to 8 computational units (nodes) with different performances measured in bcrypt hash computations per second [H/s]. In the cluster, nodes are low-power heterogeneous embedded systems with programmable logic containing specialized hash computation accelerators. In the experiments, we used a combination of Xilinx Zynq-series SoC boards and ZTEX 1.15y board which was initially used as a bitcoin miner. Zynq based nodes use the improved version of our custom bcrypt accelerator, which executes the most costly parts of the bcrypt hash computation in programmable logic. The cluster was formed around the famous open-source password cracking software package John the Ripper (abbr. JtR). On the communication layer, we used Message Passing Interface (MPI)library with a standard Ethernet network connecting the nodes. To mitigate the different performances among the cluster nodes and to balance the load, we developed and implemented password candidate distribution scheme based on the passwords' probability distribution, i.e. the order of appearance in the dictionary. We tested individual nodes and the cluster as a whole, trying different combinations of nodes and evaluating our distribution scheme for password candidates. We also compared our cluster with various GPU implementations in terms of performance, energy-efficiency, and price-efficiency. We show that our solution outperforms other platforms such as high-end GPUs, by a factor of at least 3 in terms of energy-efficiency and thus producing less overall cost of password attack than other platforms. In terms of the total operational costs, our cluster pays off after 4500 cracked passwords for a bcrypt hash with cost parameter 12, which makes it more appealing for real-world password-based system attacks. We also demonstrate the scalability of our cCc cluster.

СLASSIFIER OF BIG DATA OBJECTS OF THE SOCIO-ECONOMIC SYSTEM

The well-known models that provide the synthesis of object classifiers are analyzed: the Grothendieck topos model, Kripke’s intuitionistic model and the model of the normal Markov algorithm, in which similar approaches to the representation of objects are highlighted. The differences in the classifiers used in them are highlighted. A categorical-theoretical model of topos has been developed, in which sets are represented by n-categories. A generalized classifier has been developed that includes the properties of the Grothendieck, Kripke and Markov classifiers. An N-scheme is synthesized to replace the g-scheme of the normal Markov algorithm. A model of n-category topos has been developed and theoretically substantiated, in which the compositions of morphisms of an object are specified by an N-scheme. The alphabet of the n-category associator is synthesized, which provides the representation of non-associative compositions of morphisms, interacting objects, taking into account the assumptions laid down in the Kripke model. A database has been developed that implements the synthesized topos model. It is noted that such a database can be implemented as a cascade of unordered containers with the computation of hash functions to quickly find and retrieve keys and values. For some keys, such a base, it is allowed to duplicate a key in containers. The database provides a synthesis of self-learning on the input sampling of artificial intelligence data and supported by control lists of objects of selection, exclusion and deletion. The normal Markov algorithm is adapted to the theoretical model of the Grothendieck topos, taking into account the assumptions of the Kripke model, by replacing the g-scheme with the synthesized N-scheme.

Exploration of Mined Block Temporarily Holding and Enforce Fork Attacks by Selfish Mining Pool in Proof-of-Work Blockchain Systems

We explored mined block temporarily holding (MBTH) with enforce fork (MBTH-EF) attacks to understand the effect of selfish mining on winning rate and fairness. The temporary holding time provides a pool additional time to begin mining the next block early. In this situation, the pool may risk losing the mined block. One method to maintain the winning probability of the held block is to ensure the holding pool sends out the mined hash values once it receives the mined block message from the other pools. We analyzed MBTH-EF attacks and evaluated the effects of the win rate on when and how long a block is held. We propose a mining competition solution that does not involve actual hash calculation. It entails using one stochastic target hash value for batch racing simulation to evaluate the holding threshold, holding periods, mining time, mining difficulty, pool sizes, and rate of fork occurrences according to the operation data of the Bitcoin system. Because the periodic adjustment of mining difficulty reduces the holding effect, we explored an intermittent MBTH-EF (iMBTH-EF) attack model and examined the effect of intermittent holding on the mining game win rate for a long-term competition. We also identified suitable attack detection methods for the future work according to the simulation results.

Analysis of frequent itemset generation based on trie data structure in Apriori algorithm

Apriori is one technique of data mining association rules that aims to extract correlations between sets of items in the transaction database. The main problem with the Apriori algorithm is the process of scanning databases repeatedly to generate itemset candidates. This research examines the combination of pruning by using the trieapproach and multi-thread implementation in three algorithms to obtain frequent itemset. Trie is a data structure in the form of an ordered tree to store a set of strings where every node in the tree contains the same prefix. The use of a full combination trie (different from frequent pattern (FP) tree using links) allows the implementation of arrays and the hash calculation to achieve the addressing of itemset combination. In this research, the measure to get the address is called Hash-node calculation used to update support value. For these three alternatives, run time processing is analyzed based on the number of itemset combinations and transaction data at a certain minimum support value. The experimental results show that an algorithm thatexploits resource capabilities by applying multi-threadperforms almost seven times betterthanan algorithm implemented in single-thread in calculating hash-node. The fastest run time of the multi-thread approach is 43 minutes with 150-itemset combinations on 100,000 transaction data.

A Bloom filter variant for Blockchain

AbstractBlockchain is a single linked list of blocks consisting of transactions identified by their hash value. Querying Blockchain, primarily searching blocks/transactions, can be treated as membership queries and efficiently evaluated by a Bloom filter (BF). The existing BFs for Blockchain, however, do not exploit the data characteristics of Blockchain and the features of modern processors, which otherwise have the potential to greatly improve their performance. This article proposes a novel Bloom filter variant for Blockchain, called Blockchain Bloom Filter (BBF), with three alternatives to improve BF. First, we reorganize BF structure to make use of Single Instruction Multiple Data provided by modern processors, which parallelize the bit‐map and bit‐test processes of BBF and accelerate element insertion and membership querying. On that basis, hash computation of BBF is parallelized, which reduce the hashing time to that of the BF in theory, where k is the number of hash functions. Second, hash functions derived from block/transaction hash are developed to reduce the complexity of hash computation and accelerate the element insertion and membership query processes of BBF. Finally, we further refine BBF structure to limit the mapping range of an element within a cache‐line in the on‐chip cache in order to increase cache efficiency. With the improvements, both theoretical analysis and extensive simulations show that the performance of BBF membership queries greatly exceeds the three state‐of‐the‐art Bloom filter variants, which enables efficient data queries and application specific analysis over Blockchain.

An Intelligent Blockchain and Software-Defined Networking-Based Evidence Collection Architecture for Cloud Environment

Cloud forensics is an extension of contemporary forensic science that guards against cybercriminals. However, consolidated data assortment and storage compromise the legitimacy of digital indication. This essay proposes an evolving modern algorithm automated forensic platform based on the blockchain idea. This proposes forensic structure design, evidence gathering, and storage on a blockchain that are peer to peer. Secure Block Verification Mechanism (SBVM) will protect unauthorised users. Secret keys are optimally produced using the cuckoo search optimization method. All data are saved and encrypted at the cloud authentication server for secrecy. Confidentiality-Based Algebraically Homomorphism, a new encryption method, is given to cryptosystem learning. Every data is assigned a block in the SDN controller, and the history is kept as metadata about data. Each block has a Secure Hash Algorithm version 3 of 512-bit hash-based tree. Our approach uses graph theory-based graph neural networks in Smart Contracts to track users’ data (GNNSC). Finally, a blockchain-based evidence graph allows for evidence analysis. The experiments were run in a cloud environment with Python and network simulator-3.30 (for software-defined network). We achieved good results in terms of evidence response time, cloud evidence insertion time, cloud evidence verification time, computational overhead, hash calculation time, key generation times, and entire overall change rate of indication using our newly deliberated forensic construction using blockchain (FAuB).

Designs for efficient low power cardinality and similarity sketches by Two-Step Hashing (TSH)

In many computing applications, there is a need to estimate different features of large data sets, such as the number of distinct elements or the similarity with other data sets. This can be efficiently implemented using probabilistic data structures known as sketches; for example, Hyperloglog is widely used for cardinality estimate and Minhash for similarity estimate. Many of these sketches rely on computing hash functions on the elements and counting the number of consecutive leading zeros (or ones) to identify the minimum (or maximum) hash value for data updating. This functionality usually accounts for the most demanding part of the sketch implementation; an efficient and low power hash computation and leading zero counting are therefore required for high performance. A novel but important observation is exploited in this paper; namely the trailing hash bits are rarely used in these sketches (e.g., the leading zeros counting stops once the first one is found, which usually occurs after a few bits). Based on this feature, a so-called Two-Step Hashing (TSH) scheme is proposed in this paper to significantly reduce the power consumption. Therefore rather than computing all hash bits (as in the traditional scheme), TSH computes a few leading hash bits in a first step and the trailing hash bits in the second step, that is only required when the leading bits are all zeros (or ones). Hence, in most cases only a fraction of the hash bits is computed, so significantly reducing the power dissipation of hash computations when they are used to check the number of consecutive leading zeros, or the minimum value. The use of TSH for sketches with both single and multiple pipeline implementations are considered; evaluation results based on HyperLogLog (HLL) as a sketch with a Cyclic Redundancy Check hash function show that TSH saves up to 75.9 % (78.7 %) power dissipation in a single (multiple) pipeline implementation.

Optimization of the Distribution of Hash Calculation Tasks Flow at a Priori Given Complexity

The increasing number of computationally intensive tasks induced by digital economy development (within the framework of implementing block-chain solutions, distributed ledgers, etc.) requires more and more computational resources. At the same time users tend to move the computational process to the cloud in order to minimize costs, and the owners of cloud services are forced to look for solutions to improve their efficiency. In this paper we consider approaches that allow optimize the use of parallel computing resources for incoming sets of resource-intensive tasks, analyze different approaches to the strategy of assigning tasks to computing resources. The results of modelling experiments are provided taking into account task distribution, parameterized by execution time limit within modeling the service level agreement with the user.