1 / 23

Survey on Improving Dynamic Web Performance

Survey on Improving Dynamic Web Performance. Guide:- Dr. G. ShanmungaSundaram (M.Tech , Ph.D), Assistant Professor, Dept of IT, SMVEC. Aswini. S M.Tech CSE 2 nd year, 3 rd semester SMVEC. Agenda. Abstract Introduction Problem Issues Existing Algorithm Proposed Algorithm

Download Presentation

Survey on Improving Dynamic Web Performance

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Survey on Improving Dynamic Web Performance Guide:- Dr. G. ShanmungaSundaram (M.Tech, Ph.D), Assistant Professor, Dept of IT, SMVEC. Aswini. S M.Tech CSE 2nd year, 3rd semester SMVEC

  2. Agenda • Abstract • Introduction • Problem Issues • Existing Algorithm • Proposed Algorithm • Proposed Techniques • Proposed Architecture – Mechanism

  3. Abstract In Today’s world, every user will prefer to use a fast web browser. In this research survey, the focus is put more on improving web performance, because the faster the web, the faster the work gets done. The depth research is being focused under proposed techniques of "web cache memory compression" and "web cache optimization" with the help of a proposed algorithm "Lempel Ziv Bit Masking Hidden Markov (LZBMHM)" which will enhance the web performance.

  4. Introduction • What is web, internet and web browser? • What are the components present in web browsers? • What are the challenges and techniques in web browsers? • Which challenge will be taken and contribute to?

  5. Problem Issues The problems analyzed in the cache memory management are the memory overload of text, images and videos while navigating or browsing through various web pages. This will lead to the increase in physical memory in task manager, which may cause problem in sudden disconnection of network and high latency. Generally, the CPU utilization must be higher than Physical memory and that will not enhance the network performance, but also the whole system performance. We are not focusing in systems area here, but the network performance is affected here. Inorder to overcome this issue, the web cache memory compression and optimization are the solutions. The web cache memory compression compress the data, images and videos which compresses megabytes of data and that will reduce the memory usage in the physical memory of the task manager and that will gradually optimizes the web. Thus, this will triple the web performance.

  6. Existing Algorithm - Introduction • The existing algorithm is Lempel Ziv Markov chain Algorithm (LZMA). • The LZMA was created by a Russian programmer Igor Pavlov from the existing works of LZ77 and LZ78, where the only difference is the inclusion of the magical “range encoding” instead of Huffman coding. • LZMA is a lossless data compression.

  7. Existing Algorithm

  8. Existing Algorithm (contd…) • Step 1: The data which needs to be encoded are passed. • Step 2: The data passes through delta encoding which are digital signals and it processes into the dictionary coding. • Step 3: Sliding window encoding which is a dictionary coding mechanism where it maintains a group of strings from the input stream as the encoding process being executed. • Step 4: Then, it gets through into range encoding where it produces a space efficient stream of bits for representing a stream of symbols and their probabilities. • Step 5: Finally, the data is being encoded. • Step 6: On decoding, the encoded data passes through range decoding, where the encoded range data stream of bits gets decoded. • Step 7: Then, the encoded dictionary data which has been decoded through statistical coding passes through sliding window decoding and gets decoded. • Step 8: Thus, the data undergoes delta decoding and gets decoded. • Step 9: Hence, the decoded data passes to destination.

  9. Existing Algorithm - Limitations • The speed of compression/decompression is not good and unsatisfactory, where it is completely behind.

  10. Proposed Algorithm - Introduction • The proposed algorithm is Lempel Ziv Bit Masking Hidden Markov (LZBMHM) algorithm. • The new techniques such as Bitmasking will be adopted. • The Hidden Markov will be used here against Markov Chain which was used in existing algorithm.

  11. BitMasking • Bitmasking is termed as the code compression based technique where it helps in achieving speed of compression and decompression. • Advantages:- • Reduction in memory size. 2. Improved speed in compression and decompression. 3. Code compression, where it reduces the size of the codes. 4. No overhead. 5. Provides dynamic power in bit streaming

  12. Hidden Markov:-

  13. Hidden Markov (contd…) Earlier, The Hidden Markov currently uses three parameters for solving the problem:- (a) The first parameter uses Forward and Backward algorithms. (b) The second parameter uses Viterbi algorithm. (c ) The third parameter uses Baum Welch algorithm. But… We will "not" be using any of these three parameters mentioned above. Instead, we will be proposing a new mechanism "hidden markov web" in our proposed algorithm "Lempel Ziv Bit Masking Hidden Markov" algorithm.

  14. Proposed Algorithm

  15. Proposed Algorithm (contd…) • Step 1: The data which needs to be encoded are passed. • Step 2: The data passes through delta encoding which are digital signals and it processes into the dictionary coding. • Step 3: Sliding window encoding which is a dictionary coding mechanism where it maintains a group of strings from the input stream as the encoding process being executed. The bitmasking encoding is used in here which helps in achieving speed of the dictionary where it is the ratio of total dictionary size, total size of fully matched words, total size of bitmasked words and total uncompressed code to the number of words to be compressed and word length. • Step 4: Then, it gets through into range encoding where it produces a space efficient stream of bits for representing a stream of symbols and their probabilities. The Hidden Markov is used where it includes the linear transform and vector quantization providing improvement in efficiency of increasing compression power helping to compress more. • Step 5: Finally, the data is being encoded. • Step 6: On decoding, the encoded data passes through range decoding, where the encoded range data stream of bits gets decoded. • Step 7: Then, the encoded dictionary data which has been decoded through statistical coding passes through sliding window decoding and gets decoded through the power of bitmasking decoding where the bitmask decoding speed is high. • Step 8: Thus, the data undergoes delta decoding and gets decoded. • Step 9: Hence, the decoded data passes to destination.

  16. Proposed Algorithm - Pseudocode • Begin • // Step 1: inputs sent • // Step 2: dictionary based • if(contents exist) • output(index(word length, next symbol from the input) • continue • else • no input found • exit • // Step 3: bitmasking • Begin • // Step I: mask patterns • Begin • SlidingMask1=1ms; SlidingMask2=2ms; FixedMask1=1mf; FixedMask2=2mf • CompressionRatio=100 • for each mask m1 in (1ms, 2ms, 1mf, 2mf) • for each mask m2 in (1ms, 2ms, 1mf, 2mf) • compress=use with <m1,m2> • if(compress < CompressionRatio)

  17. Contd… • then CompressionRatio=compress • SlidingMask1=m1; SlidingMask2=m2 • endif • endfor • endfor • return SlidingMask1, SlidingMask2 • End • // Step II: bitmask dictionary • Begin • // Step a: Graph representation • G=(V,E) • for each node(V) is an unique number • where edge(E) represents bitmask that matches nodes • endfor • // Step b: Allocating to nodes and edges • /* Frequency are bit savings present in nodes */ • /* Masks are bit savings from the edges */ • // Step c: Calculating the bit savings distribution of all node.

  18. Contd… • // Step d: Selection of the best node N • // Step e: Remove N from G and insert into dictionary • for each node N1 in G that is connected to N • if(N1 < threshold) • then remove N1 from G • endif • endfor • // Step f: Repeat steps c to e until dictionary gets full or G gets empty • return dictionary • End • End • // Step 4: Compressed Code • // Conversion into compressed code from linear transformation, vector quantization and hidden markov • End

  19. Proposed Algorithm - Advantages • The speed of compression/decompression is highly improved. • The compression of text and images through encoding and decoding is done in faster way with higher compression.

  20. Proposed Techniques • Web Cache Memory Compression • Web Cache Optimization • These two techniques adopts the proposed algorithm LZBMHM.

  21. Proposed Mechanism - Architecture

  22. Conclusion Hence, an efficient “Lempel Ziv Bit Masking Hidden Markov” algorithm is proposed and it will be used along with the proposed techniques of web cache memory compression and web cache optimization. Thus, these mechanisms will enhance web performance.

  23. References

More Related