<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Arjun Thakur on Medium]]></title>
        <description><![CDATA[Stories by Arjun Thakur on Medium]]></description>
        <link>https://medium.com/@arjun.thakur20?source=rss-295856387dff------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*A5gmcvRXXOphzIXQ</url>
            <title>Stories by Arjun Thakur on Medium</title>
            <link>https://medium.com/@arjun.thakur20?source=rss-295856387dff------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 23:13:02 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@arjun.thakur20/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Application Layer Protocols]]></title>
            <link>https://medium.com/@arjun.thakur20/application-layer-protocols-9af50cbe23ff?source=rss-295856387dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/9af50cbe23ff</guid>
            <category><![CDATA[computer-network]]></category>
            <category><![CDATA[protocol]]></category>
            <dc:creator><![CDATA[Arjun Thakur]]></dc:creator>
            <pubDate>Thu, 24 Nov 2022 17:30:01 GMT</pubDate>
            <atom:updated>2022-12-02T05:53:03.508Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Authors: Arjun Thakur, Shubham Shah, Shravani Nalbalwar, Sankalp Chordia</em></p><h3><strong>Application Layer</strong></h3><p>In the Application Layer, the shared communications protocols and interface techniques that hosts in networks utilize, which is an abstraction layer. Since it is the layer closest to the user, it suggests that both the application layer and the user may directly interact with the software.</p><p>The communication between application processes (clients and servers) operating on various end systems is governed by application layer protocols. Because it is the top layer, it does not serve the lower levels. The Transport layer and any levels behind it are used by the Application layer to communicate with or transfer data to a remote host.</p><p>The application layer provides the following functions:-</p><ul><li>In order for software to communicate, receive, and present data to users in a meaningful fashion, the Application Layer offers protocols.</li><li>Users can log on as a remote host using this layer.</li><li>The application layer offers users a variety of tools, including a storage facility, to forward numerous emails.</li><li>Access to network resources is made possible through this layer, which serves as a window for users and application activities.</li><li>Host initialization and remote login to hosts are often carried out by this application layer.</li><li>In place of memorizing it or viewing it in binary format, this layer visualizes data to help people understand it (1s or 0s).</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/856/0*6rjCkf5y1lZiOI82" /><figcaption>Source: Bitlearners.com</figcaption></figure><h3><strong><em>Features Provided by Application Layer Protocols</em></strong></h3><p>In the OSI model, the application layer often primarily serves as an interface for interacting with host-based and user-facing applications. Application layer protocols are implemented identically on the source host and destination host to enable smooth communication.</p><p>The features offered by application layer protocols include the following:</p><ul><li>The communication procedure between the two parties is defined by the application layer protocol.</li><li>The type of message being transmitted or received from any side is specified by these protocols (either source host or destination host).</li><li>These protocols also specify the message’s fundamental syntax when it is forwarded or retrieved.</li><li>These protocols specify the message-sending procedure and the anticipated response.</li><li>Additionally, these protocols outline how to communicate with the Transport level.</li></ul><h3>Types of Protocols</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/389/1*Ugde5tJaW3SFRrcYTdUeFQ.jpeg" /><figcaption>Soruce: cpltutorials.com</figcaption></figure><blockquote><strong>TELNET</strong></blockquote><p>Teletype Network is known as TELNET. An individual computer can connect to a nearby computer using this particular protocol type. It serves as a common TCP/IP protocol for the ISO-provided virtual terminal service. The local computer is the one that initiates a connection. Remote computer is a computer that is connected to, or one that accepts the connection. when the local and distant computers first make contact with one another. Whatever is being done on the distant computer when you are using telnet will be seen on the local computer. Telnet uses a client/server model to operate. The distant computers utilize the telnet server application, while the local computer uses the telnet client program.</p><p>The command format for TELNET is specified below:</p><p>The majority of telnet implementations run in one of the three main modes:</p><p><strong>Default Mode:</strong></p><ul><li>This mode is utilized if no other modes are activated.</li><li>In this mode, the client is in charge of echoing.</li><li>In this mode, when a character is entered, the client displays it on the screen but does not send it until the entire line has been entered.</li></ul><p><strong>Character Mode:</strong></p><ul><li>In this mode, the client sends every character they type to the server.</li><li>In this sort of mode, the server often echoes characters back for the client to see on their screen.</li></ul><p><strong>Line Mode:</strong></p><ul><li>In this line, Client-side line editing is used for things like echoing and character erasure.</li><li>The entire line will be sent by the client to the server.</li></ul><blockquote><strong>FTP</strong></blockquote><p>File Transfer Protocol(FTP) is an application layer protocol that moves files between local and remote file systems. It runs on the top of TCP. To transfer a file, 2 TCP connections are used by FTP in parallel: control connection and data connection.</p><ul><li>For sending control information like user identification, password, commands to change the remote directories, etc., FTP makes use of a control connection. The control connection is initiated on port number 21.</li><li>For sending the actual file, FTP makes use of a data connection. A data connection is initiated on port number 20.</li><li>FTP sends the control information out-of-band as it uses a separate control connection. Some protocols send their request and response header lines and the data in the same TCP connection.</li></ul><p>The client opens a control TCP connection with the server-side when an FTP session is begun between a client and a server. Over this, the client transmits control information. In response, the server opens a data connection to the client-side. Over a single data connection, just one file may be transmitted. However, the control connection is active the entire time the user is logged in. Since HTTP is stateless, it is not required to maintain any user state. But throughout the session, FTP must keep track of its user’s condition.</p><p>Data Structures: FTP allows three types of data structures :</p><ul><li>File Structure — In file structure, there is no internal structure and the file is considered to be a continuous sequence of data bytes.</li><li>Record Structure — In record structure, the file is made up of sequential records.</li><li>Page Structure — In page structure, the file is made up of independent indexed pages.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/984/0*o8RJ_l9hINTaAUUi" /><figcaption>Source: Baeldung.com</figcaption></figure><blockquote><strong>HTTP</strong></blockquote><p>Hyper Text Transfer Protocol is known as HTTP. The World Wide Web (WWW) is built on it . The client-server architecture underlies HTTP. For the transmission of hypermedia documents like HTML, this protocol is utilised.</p><p>Although this protocol was created specifically for interactions between web browsers and web servers, it can also be used for a number of other things. Because HTTP is a stateless protocol (a network protocol in which a client submits requests to a server and the server responds based on the given state), the server is not obligated to keep track of the requests from earlier clients. 80 is the port used by HTTP.</p><p>Following is a list of features controllable using HTTP:</p><ul><li><strong>Caching:</strong> Clients and proxies can get instructions from the server on what to cache and for how long. The saved document can be disregarded by intermediate cache proxies at the client’s request.</li><li><strong>Sessions: </strong>The user can connect requests with the server’s status via HTTP cookies. Despite the fact that basic HTTP is a stateless protocol, this creates sessions. This is helpful for any website that allows users to customise the output, not just e-commerce shopping baskets.</li><li><strong>Proxy and Tunneling: </strong>The genuine IP address of servers or clients is frequently hidden from other computers via intranets. In order to get beyond this network barrier, HTTP queries then go through proxies.</li></ul><blockquote><strong>TFTP</strong></blockquote><p>The Trivial File Transfer Protocol is referred to as TFTP. Some applications don’t require or can’t afford TCP’s full capabilities or complexity. TFTP enables a simple structure that doesn’t call for costly client-server communication. TFTP does not support authentications and restricts operations to straightforward file transfers.</p><p>Using TFTP has the advantage of allowing bootstrapping programmes to use the same underlying TCP/IP protocols that the operating system does once it begins to run. So there is a chance that a device may start up from a server on a different physical network.</p><p>Features of TFTP:</p><p>The following are TFTP’s primary features:</p><ul><li>The TFTP server employs the well-known UDP port number 69 and is based on the client-server model. TFTP does not provide authentication and is an unsafe protocol. The idle RQ (stop and wait) error recovery technique is part of TFTP.</li><li>TFTP data units each have a unique sequence number.</li><li>Each data unit is recognised independently. The next data unit is transferred after receiving the acknowledgment.</li><li>Retransmission is used for error recovery after a timeout. The TFTP protocol employs an exponential back-off mechanism with adjustable timeout.</li></ul><p>There are 4 types of TFTP Messages namely read request, write request, data and acknowledgement.</p><blockquote><strong>NFS</strong></blockquote><p>A networking protocol for widespread file sharing is called Network File System (NFS). A file system specifies how data is saved and retrieved in the form of files from storage devices including tape drives, solid-state drives, and hard disc drives. The method for storing and retrieving files from storage devices via networks is outlined by the NFS network file sharing protocol.</p><p>There are three NFS versions for us, with each having different standards for how it will operate, as follows:-</p><ul><li><strong>NFS version 2 (NFSv2):</strong> is the most well-supported and earliest format you may utilise. It uses an IP network and the User Datagram Protocol, or UDP, to function. A reliable network connection is possible thanks to the IP network.</li><li><strong>NFS version 3 (NFSv3):</strong> enables asynchronous writes, which give the server control over the ideal synchronisation rules. Before a command to commit to managing the data is established, the data will be synced. Compared to NFSv2, the architecture produces better buffering.</li><li><strong>NFS version 4 (NFSv4):</strong> It is the most latest NFS protocol that is available for use. Both the internet and firewalls are compatible with it. It can run in more locations because it doesn’t need a rpcbind service. In this NFS format, the Transmission Control Protocol, or TCP, operates. the TCP connections between an IP and an application.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/851/0*gX9tONowx2tnPwZ-" /><figcaption>Source: GeeksForGeeks.com</figcaption></figure><blockquote><strong>SMTF</strong></blockquote><p>Simple Mail Transfer Protocol is known as SMTP. One user can send electronic mail to another user using it. SMTP runs on ports 25 and 587. Setting up communication rules between servers is the major usage of SMTP.</p><p>The servers can identify themselves and state the type of communication they are attempting to carry out. They also provide a method for dealing with issues like a wrong email address.</p><blockquote><strong>LPD</strong></blockquote><p>Line Printer Daemon, often known as LPD, is a common TCP/IP daemon used on UNIX networks for accepting and spooling print jobs on a print server. Either the print device is directly linked to the Line Printer Daemon (LPD) print server or it is a network print device that supports LPD. The specifications of LPD are given by RFC 1179.</p><p>Benefits of the LPR/LPD Protocol:</p><ul><li>It’s easy to implement, which suggests that it’s a solid strategy.</li><li>Supports ordering, queueing, and spooling; manages system and/or network disruptions.</li><li>The bulk of UNIX and other systems that offer LPD services are fundamentally compatible with using the LPD/LPR protocol.</li><li>Information in the control file is available for use in a variety of implementations.</li></ul><blockquote><strong>DNS</strong></blockquote><p>DNS stands for Domain Name Service. It is an application layer protocol that specifies how message passing between application processes running on various systems works. A directory service called DNS establishes a connection between a host’s name and address on a network. The functionality of the internet depends on DNS. A service called DNS converts domain names into IP addresses. This enables network users to locate other hosts by using friendly names rather than their memory of IP addresses.</p><p>Generic domains, country domains, and inverse domains are the three divisions of the domain name space.</p><p><strong>Working of DNS:</strong></p><ul><li>Client-server network communication is facilitated by DNS. While DNS servers respond to queries from clients, DNS clients make requests to the server.</li><li>Reverse DNS lookups are client requests in which an IP address is transformed into a name, whereas forward DNS lookups are client requests in which a name is converted into an IP address.</li><li>In order to keep the names of all hosts that are accessible via the internet, DNS creates a distributed database.</li><li>When a client, such as a web browser, submits a request with a hostname, a piece of software, such as a DNS resolver, submits a request to the DNS server to acquire the IP address of the hostname.The request is sent to another DNS server if the DNS server does not have the IP address linked to a hostname. If the resolver receives an IP address, it completes the request through the internet protocol.</li></ul><blockquote><strong>DHCP</strong></blockquote><p>Dynamic Host Configuration Protocol (DHCP) is a client/server protocol which automatically assigns an Internet Protocol (IP) host with its IP address and other necessary configuration data like the subnet mask and default gateway. It is an application layer protocol which is used to provide:</p><ul><li>Subnet Mask (Option 1 — e.g., 255.255.255.0)</li><li>Router Address (Option 3 — e.g., 192.168.1.1)</li><li>DNS Address (Option 6 — e.g., 8.8.8.8)</li><li>Vendor Class Identifier (Option 43 — e.g., ‘unifi’ = 192.168.1.9 #where unifi = controller)</li></ul><p>DHCP uses the discovery, offer, request, and ACK protocols and is based on a client-server architecture.</p><p>DHCP port numbers are 67 for the server and 68 for the client. It employs UDP services and is a client-server protocol. A pool of addresses is used to assign IP addresses. The DHCP connection process, also known as the DORA process, involves the client and server exchanging 8 DHCP messages, though primarily 4 DHCP messages. Without DHCP, configuring IP addresses for new machines or those moving from one subnet to another requires manual effort. This entire procedure is mechanised and centralised with DHCP.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*UMuYSVoHqrV-J1GX" /><figcaption>Source: Baeldung.com</figcaption></figure><h3>Conclusion</h3><p>In both the OSI and TCP/IP models, the Application Layer is the top layer. Since the application layer is the one that is closest to the end user, both the application layer and the user may have direct communication with the software application.</p><p>Some of the protocols used at the application layer are TELNET, File Transfer Protocol (FTP), Simple Mail Transfer Protocol (SMTP), Domain Name System (DNS), and Simple File Transfer Protocol (TFTP). The receiver is identified, reachable, and prepared to accept data, for example, thanks to the services provided by the application layer. It also guarantees the network has the required communication interfaces. Users can log on to the remote host via the application layer, which also offers features like email, file transfer, results distribution, directory services, and network resources.</p><p><em>Originally published at </em><a href="https://arjunt13.github.io/CN_Blog/"><em>https://arjunt13.github.io</em></a><em> on November 24, 2022.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9af50cbe23ff" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Supercomputers]]></title>
            <link>https://medium.com/@arjun.thakur20/supercomputers-e08625a4776?source=rss-295856387dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/e08625a4776</guid>
            <category><![CDATA[fugaku]]></category>
            <category><![CDATA[cpu]]></category>
            <category><![CDATA[supercomputer]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[supercomputing]]></category>
            <dc:creator><![CDATA[Arjun Thakur]]></dc:creator>
            <pubDate>Sun, 01 May 2022 17:30:00 GMT</pubDate>
            <atom:updated>2022-06-07T18:12:13.833Z</atom:updated>
            <content:encoded><![CDATA[<blockquote><strong>Authors:</strong> Shubham Shah, Shruti Singh, Akash Sinha, Vishal Sirvi, Shreyansh Srivastava and Arjun Thakur.</blockquote><h3>What makes a Computer “Super”?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*mD1anYM0EcYXWzlD" /></figure><p>Faster processing speeds, extra memory, and super-sized storage capacity are what make supercomputers the tools of choice for many researchers. The devices we keep at home and in our pockets are pretty advanced compared to the technology of the past, but they can’t hold a candle to the raw power of a supercomputer.</p><p>A supercomputer is a computer with a high level of performance as compared to a general-purpose computer. The performance of a supercomputer is commonly measured in floating-point operations per second (FLOPS) instead of million instructions per second (MIPS). It performs at or near the highest operational rate for computers.</p><p>Traditionally, supercomputers have been used for scientific and engineering applications that must handle massive databases, do a great amount of computation or both. Advances like multicore processors and general-purpose graphics processing units have enabled powerful machines that could be called desktop supercomputers or GPU supercomputers. Supercomputers have very high storage capacity and can store thousands of times more data than any average personal computer. The supercomputer can process billions of instructions in a second. Modern supercomputers have thousands of processors and are used for complex computations.</p><p>Processing power is the main difference between supercomputers and general-purpose computer systems. A supercomputer can perform 100 PFLOPS. A typical general-purpose computer can only perform hundreds of gigaflops to tens of teraflops. Supercomputers consume lots of power. As a result, they generate so much heat that they need to be stored in cooling systems.Both supercomputers and general-purpose computers differ from quantum computers, which operate based on the principles of quantum physics.</p><h3>The 3 fastest supercomputers in the world</h3><h4>Fugaku (Japan)</h4><p>Jointly developed by <a href="https://www.riken.jp/en/">RIKEN</a> and Fujitsu, Japan’s Fugaku is the new number one fastest supercomputer in the world. Japan has not had a system take the top spot since June 2011 when Fugaku’s predecessor, <a href="https://www.top500.org/lists/top500/2011/06/highlights/">the K computer</a>, debuted in first place. Boasting nearly 7.3 million cores and a speed of 415.5 petaFLOPS, Fugaku far outperforms previous #1 Summit’s 148.6 PetaFLOPS, bringing HPC technology one step closer to the promised exascale era.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*93FFLQGhJVE-JkIPh8HeGw.jpeg" /><figcaption><a href="https://www.japantimes.co.jp/news/2021/03/09/national/science-health/fugaku-supercomputer-covid19-research/">FUGAKU SUPERMCOMPUTER</a></figcaption></figure><h4>Summit (US)</h4><p>Poised at the Summit. The US Department of Energy’s Oak Ridge National Laboratory (ORNL) in Tennessee slips to #2 after two years in first place. Courtesy ORNL.As part of the US Department of Energy’s renewed commitment to supercomputing power, Oak Ridge National Laboratory’s (ORNL) Summit first claimed the #1 spot in June 2018, taking the top rank from China for the first time in 6 years.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Tmpqe2p1vX4ppJCPZjFHTg.jpeg" /><figcaption><a href="https://wire19.com/summit-is-now-the-most-powerful-computer-on-planet/">SUMMIT SUPERCOMPUTER</a></figcaption></figure><h4>Sierra (US)</h4><p>Lawrence Livermore National Laboratory’s (LLNL) Sierra initially debuted at #3 on the June 2018 list with 71.6 petaFLOPS. Optimization later pushed the processing speed on its 1,572,480 cores to 94.6 petaFLOPS, earning it the #2 spot in November 2018. However, the ascension of a new number one in June 2020 pushes Sierra back to third position.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*pB9fkoUs9iC7jBWvcS6ZWA.jpeg" /><figcaption><a href="https://www.rawpixel.com/image/3325557/free-photo-image-server-room-supercomputer">SIERRA SUPERCOMPUTER</a></figcaption></figure><h3><strong>Evolution of Supercomputers</strong></h3><p>The United States, China, the European Union and Japan are already competing to develop supercomputers that are faster, more powerful, and technologically advanced. However, the US marked a big stride in the history of supercomputers by introducing CDC (Control Data Corporation) 6600, in 1964.</p><h4><strong>Early Supercomputers</strong></h4><p>From the mid-1980s through the early 1990s, Cray dominated supercomputers. The Cray Y-MP, with up to eight 32-bit vector processors operating at 167MHz, was introduced by Cray Research in 1988. It was the first supercomputer to sustain more than 1GFLOPS (109 floating point operations per second) and provided SRAM main memory options of 128, 256, or 512MB.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4uol3UL6_eOQPySK6Q7Q8Q.jpeg" /><figcaption><a href="https://en.wikipedia.org/wiki/Cray_Y-MP">Cray Y-MP Supercomputer</a></figcaption></figure><p>The X-MP, a forerunner of the Y-MP, was marketed to a West German nuclear research organization worth $11.4 million in 1981, or $32.6 million in 2020 dollars. Cray may have dominated the supercomputing business in the early 1990s, but they were far from alone. NEC’s SX line of vector supercomputers was named after it. The first two NEC models, the SX-1 and SX-2, were introduced in 1985. Both systems have up to 256MB of main memory. The SX-2 was the first supercomputer to surpass 1 GFLOPS, according to claims. It has four high-performance vector operation pipelines, each with up to 16 arithmetic units capable of performing multiple/parallel operations. The NEC SX-1 cost about half as much as the SX-2 and offered about half the performance.</p><h4><strong>Supercomputers in the Latter 1990s era</strong></h4><p>NEC introduced the SX-5 system in 1998. It has the ability to attain a performance of 4 TFLOPS. Each system node contains up to 128GB of primary memory and 16 CPUs. Up to 32 nodes might be linked together for a total memory capacity of 4,096GB.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vgL1PfTsarrdT2WwGj1Jsg.jpeg" /><figcaption><a href="https://en.wikipedia.org/wiki/NEC_SX">NEC SX Vector Supercomputer</a></figcaption></figure><p>SGI introduced the SGI Origin 2000, a large-scale HPC system, in 1996. It could handle CPUs with a performance range of 2 to 128 MIPS. These were MIPS R10000 processors, which started at 180MHz and increased to 300 and 400MHz over time. SGI had the advantage of having non-supercomputer CPUs, such as the Origin 2000. They were also used in SGI workstations. Both “traditional” supercomputers and PC systems had a key decade in the 1990s. Both had made rapid progress, setting a course for the early 2000s.</p><h3><strong>Evolution of Indian Supercomputer</strong></h3><p>Dr. Vijay Bhatkar is known as the “Father of Indian Supercomputers’ ‘, he introduced the Indian PARAM Supercomputer. In November 1987, the Centre for Development of Advanced Computing (C-DAC) obtained a three-year budget of Rs 375 million to create supercomputers with 1000Mflops (1 Gflop). Three C-DAC flights revealed the “PARAM’’ (Parallel Machine) supercomputer family.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/728/1*savfVeCqh9xgbAJJRwwZ4Q.png" /><figcaption><a href="https://cse.iitk.ac.in/users/pmalakar/acmws/Supercomputing_in_India.pdf">Development of PARAM supercomputer over the years</a></figcaption></figure><h4><strong>PARAM 8000</strong></h4><p>In August 1991, the PARAM 8000, a 64-node system, was introduced as the first machine designed from the ground up. This 256-node system has a theoretical performance of 1 Gflops, but in actuality, it only has roughly 100–200Mflops. The PARAM 8000 was built on a distributed memory MIMD architecture with a reprogrammable connectivity network.</p><h4><strong>PARAM 8600</strong></h4><p>PARAM 8600 was released in 1992 as an upgrade to PARAM 8000. The Intel i860 CPU was chosen by C-DAC as a way to increase power. Each 8600 cluster was equivalent to four PARAM 8000 clusters in terms of processing power.</p><h4><strong>PARAM 9000</strong></h4><p>The PARAM 9000 was initially shown in 1994, and it was created to combine cluster and massively parallel processing computer tasks. Employing the Clos network design, this system expanded up to 200 CPUs using 32–40 processors. The SuperSPARC II processor was used in the PARAM 9000/SS; the UltraSPARC processor was used in the PARAM 9000/US; and the DEC Alpha was used in the PARAM 9000/AA.</p><h4><strong>PARAM 10000</strong></h4><p>In 1998, the PARAM 10000 was revealed. This supercomputer featured many independent nodes, each of which was based on a Sun Enterprise 250 server with two 400 Mhz UltraSPARC II processors. This system’s top speed was 6.4 Gflops. This would have 160 CPUs and 100 Gflops of performance, readily expandable to the Tflop range.</p><p>Later on, PARAM Padma, PARAM Yuva, PARAM Brahma, PARAM Ishan and PARAM Siddhi-AI were introduced which were known to be technologically advanced successors of each other and have high computing power. The PARAM Siddhi-AI supercomputer is India’s fastest, with a peak Rpeak of 5.267 Pflops and a maximum Rmax of 4.6 Pflops (Sustained). The AI assists research in innovative materials, combinatorial chemistry, and astrophysics, as well as universal healthcare, disaster prevention, and the Covid-19 application, by allowing for faster simulations, diagnostic imaging, and genome sequencing. In November 2020, PARAM Siddhi-AI was ranked 63rd amongst world’s best supercomputers. NVIDIA DGX SuperPOD networking is used, as well as C-own DAC’s HPC-AI engine, software frameworks, and cloud platform.</p><h3><strong>Different types of Supercomputers</strong></h3><p>There are two types of supercomputers which evolve gradually</p><ul><li>General Purpose Supercomputers.</li><li>Special Purpose Supercomputers.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/580/1*foV6hXoiVO_Jn7-gxmQNKw.jpeg" /><figcaption><a href="https://towardsdatascience.com/super-computing-168be247335b">SUPERCOMPUTER</a></figcaption></figure><h4>1. General-purpose supercomputer</h4><p>Further General Purpose supercomputers can be classified as:</p><ul><li><strong>Vector processing computers</strong>: Vectors or array processors are used in a vector processing computer. These processors are similar to CPUs that can assist in performing mathematical operations on a large number of data pieces.</li><li><strong>Commodity computers</strong>: Computers that are a collection of commodity computers or conventional PCs connected to a local area network with low latency and high bandwidth.</li><li><strong>Tightly connected cluster computers</strong>: A cluster supercomputer is a collection of closely linked computers that operate as a single entity.</li></ul><h4>2. Special purpose supercomputers</h4><p>Supercomputers that are developed specifically to accomplish a specific task or aim are known as special purpose computers. They frequently employ Application-Specific Integrated Circuits (ASICs), which outperform general-purpose supercomputers. These supercomputers are used for brute-force codebreaking and astrophysical computing while their hardware architecture is designed to solve a single task.</p><h3><strong>Supercomputers and Artificial Intelligence</strong></h3><p>Supercomputers have the potential to boost the speed of artificial intelligence systems.</p><p>However, because super-computing is commonly used for data analysis and scientific activities like processing massive amounts of data to solve clinical, environmental, infrastructure, and a variety of other scientific problems, few members of the general public have a thorough understanding of how advanced technology affects their lives.</p><p>Supercomputers have incredible processing speed, allowing them to convert simple data into useful data in seconds, minutes, or days, rather than the years or decades it would take if done manually.</p><p>The world’s new fastest supercomputer, according to Data Center Knowledge, is called Fugaku and is powered by Fujitsu’s 48-core A64FX SoC. The RIKEN Center for Computational Science (R-CCS) in Kobe, Japan, has it installed. The non-profit Top500 organization announced on June 22, 2020 that it outperforms the previous performance champion, the US Department of Energy’s Summit system, which topped the Top500 list twice in a row.</p><p>Artificial intelligence (AI) is putting machine-assisted capabilities to the test. It has the potential to boost the speed with which machines perform human-like tasks. Because of its automation and advanced analytics, AI is becoming more important. Using the power of machine learning, deep learning, and natural language processing, AI is providing tremendous benefits to organizations. Artificial intelligence (AI) helps companies capitalize on new digital industry trends. Artificial intelligence will benefit individuals, markets, and society as a whole.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/634/1*kC540X0kEKpSp48nNAIcJA.jpeg" /><figcaption><a href="https://www.google.com/url?sa=i&amp;url=https%3A%2F%2Fwww.dailymail.co.uk%2Fsciencetech%2Farticle-10435181%2FMeta-building-worlds-fastest-AI-SUPERCOMPUTER.html&amp;psig=AOvVaw06xaqlVtd3sOdk1nn9M40p&amp;ust=1654711458121000&amp;source=images&amp;cd=vfe&amp;ved=0CA0QjhxqFwoTCKDWs9n2m_gCFQAAAAAdAAAAABBB">Meta’s AI Supercomputer</a></figcaption></figure><p>Although supercomputers have long been required in fields like physics and space science, the increased use of artificial intelligence and machine learning has resulted in a surge in demand for supercomputers capable of performing quadrillions of calculations per second. In reality, exascale supercomputers, the next generation of supercomputers, are improving efficiency in these areas.</p><p>Supercomputers, or, to put it another way, machines with accelerated hardware, are deserving of being used to increase the speed of artificial intelligence systems. Because of its improved pace and ability, it can train faster on larger, more detailed sets, as well as more oriented and deeper training sets.</p><p>According to AIM, Microsoft announced last year that it had developed a supercomputer in collaboration with OpenAI and would be hosted in Azure Cloud. The supercomputer, according to the company, has over 2,85,000 CPU cores, 10,000 GPUs, and 400Gbps network connectivity for each GPU server, allowing it to train various artificial intelligence models.</p><h3><strong>Supercomputing Challenges</strong></h3><p>Most of the challenges and issues related to supercomputing are because of the hardware and software interrelationships and their algorithms. However there are some other challenges as well such as movement of data , fault tolerance, power consumption, and extreme parallelism. In this section we will be talking about all these challenges and issues in brief.</p><h4><strong>Speed of data moment</strong></h4><p>This challenge basically refers to the lack of available technology which are mainly used to fetch and retrieve data at sufficiently high rates. For faster computation, fast execution speed and data transfer rate is very important and that too has to be compatible with the given power packet which gives rise to the other challenge about which we will discuss further. Actually in most of the cases it has been found that the time and energy required for data movement from main memory to the processing unit and from processing unit to main memory again is greater in magnitude than the energy and time required for the actual computation and operation.</p><h4><strong>Fault Tolerance</strong></h4><p>Fault tolerance is the ability of a system to continue the given computational task without being affected by any available faults in the program. Incase of supercomputing where a huge number of components and peripherals are being interfaced with larger systems along with advanced technologies being operated at specific and limited voltage levels, providing this feature is yet another challenge.</p><h4><strong>Power Consumption</strong></h4><p>As discussed earlier, power consumption rates along with the speed of data transfer are the most basic and yet challenging issues faced in the field of supercomputing. As per findings, average power consumption of an exaFLOP built with the existing available technologies would be more than 600 megaWatt. This will not only require a more powerful source of energy but also increase the cost of computing.</p><h4><strong>Level of Parallelism</strong></h4><p>For faster execution, greater no of threads along with multilayer parallelism is needed. Infact, if the computation rate is to be as fast as 1 exaFLOP, the requirement is of 1 billion floating point units, where each unit is performing 1 billion calculations per second. As far as parallelism is concerned, the requirement is parallel processing of billion threads being solving a small segment of a single problem.</p><h3><strong>Future of Supercomputers</strong></h3><p>As we all know, the first supercomputer appeared in the 1960s, principally developed by Control Data Corporation. This business led the market until its co-founder Seymour Cray departed to create his own company, Cray Research, in the 1970s. He subsequently dominated the market with his innovative ideas, dominating supercomputing for several years. A number of smaller rivals joined the market in the 1980s, but many vanished after the mid-1990s supercomputer market meltdown.</p><p>After a quick overview of what supercomputers were and when they first appeared, let us attempt to predict their future. For this, we will need to refer to a supercomputing performance metric known as floating point operations per second (FLOPS). FLOPS, or floating-point operations per second, is the unit of measurement used to quantify a supercomputer’s performance capabilities.</p><p>Sharing some FLOPS statistics from previous generations of supercomputers. For example, Intel’s ASCI Red supercomputer was the first to reach one teraFLOPS (a trillion computations per second) in 1997. It had a peak performance roughly 800 times faster than the Cray-2, which had been introduced only 12 years before in 1985. In 2008, IBM’s Roadrunner reached one petaFLOPS (1,000 trillion computations per second), which is around 1,000 times faster than ASCI Red.</p><p>Supercomputers have continued to advance by orders of magnitude since then, but at a slightly slower rate especially from the second half of 2013, according to the TOP500 list kept by academics in the United States (which employs the LINPACK benchmark) and on which the graph stated below is based on. Nonetheless, the most powerful machines will soon surpass the next milestone: exaFLOPS. In other words, 1,000,000,000,000,000,000,000,000 floating point operations per second. Several candidates are in the wings, and one of them may win the championship in the near future.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/612/0*yeyyKz-P0N5dvI5O" /><figcaption><a href="https://www.futuretimeline.net/data-trends/images/supercomputer-future-predictions.jpg">Supercomputer Future Prediction</a></figcaption></figure><p>Some future supercomputers are already in the works. For example, Oak Ridge National Laboratory in Tennessee is creating a $600 million machine named Frontier, which is expected to arrive by the end of 2021 and has a goal capability of 1.5 exaFLOPS. Meanwhile, on the suburbs of Chicago, the Argonne National Laboratory is creating Aurora, a system created by Intel and Cray that will have more than 200 cabinets with a combined 1 exaFLOPS. This was supposed to be the world’s first exaFLOPS computer, but owing to chip production delays with Intel, it will now arrive after Frontier.</p><p>Isn’t the race to become a worldwide supercomputing power fierce? So, let’s see who is currently at the top of the world rankings. China now leads the TOP500 list, with 188 computers (compared to 122 for the United States), although it only has two machines in the top ten. Japan already boasts the world’s fastest supercomputer, Fugaku, which has a speed of 442 petaFLOPS. As a result, the first exaFLOPS machine will require at least 2.3 times that speed.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/904/0*MJv6P3AxdLGo9npo" /><figcaption><a href="https://www.datacenterknowledge.com/sites/datacenterknowledge.com/files/styles/article_featured_retina/public/supercomputer%20fugaku.jpg?itok=WBLNzL95">FUGAKU SUPERCOMPUTER</a></figcaption></figure><p>Looking even farther forward, we reach yottaFLOPS, a computing milestone that might occur in the early 2050s. A supercomputer with yottaFLOPS would be 1,000 times faster than a zettaFLOPS machine and a million times faster than an exaFLOPS machine. To put this in context, a yottaFLOPS supercomputer could finish a large-scale simulation that would have taken six months to perform in 2021 in only 15 seconds.</p><p>What can such speed and computational power get us? The uses of such a machine might open up a whole new world of possibilities, which are part of the ‘Singularity’ predicted by many futurists. We can only speculate on the kinds of game-changing discoveries that will result from this technology and its exponentially increased possibilities. Perhaps we will be able to reconstruct the early moments of the Big Bang in minute detail. Or perhaps a new material will be discovered at the nanoscale that is far more sophisticated than anything now available.</p><p>Along with the development of the first super AI, with a supercomputer as its brain, is an even more fascinating (and terrifying) potential. According to one scenario foreseen by futurists such as Life 3.0 author Max Tegmark, this organism may attempt to escape its confinement and spread over the web. It might then download itself onto robot bodies, resembling the premise of a science fiction film. It remains to be seen if this will be beneficial or bad for humanity.</p><blockquote><strong>Give a couple of claps if you found the article useful. Thank You and Happy Learning !!!</strong></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e08625a4776" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>