Sergey Lebedev was born on November 2nd 1902, into the family of Alexey, a talented educator, and Anastasia, who was a teacher of junior grade levels at a national public school in Nizhny Novgorod. His father, Alexey Lebedev, was known as the author of the famous "ABC" and the "Dictionary of odd words" in Russia. Shortly after the revolution, he was invited by Anatoly Lunacharsky to work as the People’s Commissar of Education in Moscow, and the Lebedevs moved to the capital. In 1921, 19-year-old Sergey passed his high school exams without attending classes and was enrolled into the Moscow Higher Technical College (named after Bauman) at the Electrical Engineering Faculty.
In 1928 he received a diploma of electrical engineering and he stayed at the college to teach. At the same time, he occupied the post of junior research fellow of the Soviet Electro Technical Institute (SETI). Soon he became head of the group and then of the electrical networks laboratory there.
In 1933, together with the famous scientist, P.S.Zhdanov, he published a monograph, "The Stability of the Parallel Operation of Electrical Systems," which was edited and reprinted in 1934. The next year, the Higher Attestation Committee gave him the title of professor. In 1939 Lebedev, not being an Associate of Science, defended his PhD dissertation which focused on his theory of the artificial stability of power systems.
Lebedev worked in Moscow for almost 20 years, during 10 of which he headed the Department of Automation, dealing with the modeling and controlling of power at generating facilities. At the beginning of the Great Patriotic War in September 1941, Sergey was evacuated with the Institute to Sverdlovsk. There he designed the stabilization system for a tank gun for targeting, which was put into production remarkably quickly. This system made tanks less vulnerable and saved the lives of many tankers, as it made it possible to aim and fire without stopping the machine. For this work, Lebedev was awarded the Order of the Red Banner and the medal, "For Valorous Labor in the Great Patriotic War of 1941-1945.” In order to develop the stabilization system, he used so-called analog elements or vacuum tubes.
In 1944 the war was still going on, but Sergey moved to Kyiv, where he was elected to become an academician of the USSR and was appointed as director of the Energy Institute at the Academy of Sciences of Ukraine.
A story about moving to Kyiv was told by his son. It turns out that Lebedev doubted whether he should accept the offer from the Ukrainian Academy of Sciences. So his wife, Alisa Lebedeva, suggested to him that he cast lots. Two folded pieces of paper with the words "Kyiv" and "Moscow" were put into a hat and tossed thoroughly. It was Kyiv.
Having accepted the offer at the Institute, where two incompatible scientific fields were developing – electrical and heat engineering – the new director decided to split these fields into separate institutions. Lebedev became the director of the Institute of Electrical Engineering of the Academy of Sciences of the USSR.
Initially, he organized a simulation and control laboratory and continued his previous research in developing devices for compounding power generators to increase the stability of power systems and to improve performance of electrical facilities. For this work, in 1950 he and L.V. Tsukernik were awarded the USSR State Prize.
The binary system was also an area of interest to the scientist. If it hadn’t been for the war, Lebedev would have started earlier on his work to create a computer using the binary system. The staff that worked with him shared this opinion. During those times there were not any sufficiently complete publications on the binary number system and on the methods of operations with binary numbers. Lebedev’s developed technique of arithmetic operations in the binary system became the theoretical basis for the construction of a digital computer, as well as Lebedev’s previously developed numerical methods for solving mathematical problems.
After moving to Kyiv, and by the end of 1948, Sergey was focused on the long-planned execution of the development of a digital computing machine. According to Lebedev, he worked out the basic principles of its construction in 1948-1949. During the first stage, Lebedev’s new creation was named the Model of the Electronic Computing Machines (MESM). In November 1950, the MESM performed its first calculation: the sum of odd numbers range of factorial and raising it to the power. In December 1951, the computer was adopted by the State Commission and put into operation.
In March 1950, Lebedev was concurrently appointed as the director of laboratory № 1 at the Moscow Institute of Precision Mechanics and Computer Technology (IPM and CT) at the Academy of Sciences. He started actively working on a large electronic computer (BESM). In 1952 he left Kyiv and moved to Moscow, passing on responsibility for the development of his specialized computer, which was designed to solve the systems of algebraic equations, to his former graduate student Z.L. Rabinovich.
According to those who worked with Sergey Lebedev in Kyiv, he was a perfect leader. He saw in great detail the developmental processes of his projects. It is important to mention that he planned MESM entirely on his own, engaging the staff only after all the necessary explanations of what and how to design it had been made.
He never raised his voice, even with those who failed badly. He treated everyone very calmly and fairly. He showed no favoritism and always acknowledged even the little successes of his employees. During the computer debugging, there was no one equal to him, as he excelled in understanding the problems and failures of the device. Sergey had his own "methods" of finding faults with precision, up to a unit. "In many blocks of the first BESM in the anode valves chains, they used, not resistance, but ferrite transformers,” says V.S. Burtsev. “Since these transformers were crafted by amateurs, they were often burnt out, releasing a peculiar acrid odor. Sergey had a remarkable sense of smell, and would sniff the system of each block to find the defective unit. He made almost no mistakes.”
The first MESM was produced in a single version. The production of machines designed at IPM and CT began in the USSR after seven years, in 1958. The BESM mass production was unable to start earlier, as the Ministry of Machine Engineering and Instrument-making Industry did not provide sufficient materials to produce the memory for the BESM. As a result, Lebedev had to build the the memory out of mercury tubes, thereby reducing the performance of the BESM by five times. That fact prevented the BESM from becoming the first Soviet computer in mass production and the most productive computer of its time in the world.
In 1958 the IPM and CT team, under the leadership of Lebedev, developed and put into production two computers: the BESM-2 (an upgraded BESM) and the M-20. In the BESM-2, they implemented random access memory on the ferrite cores. They also widely used semiconductor diodes as well as improved design (building block). This computer could already solve important problems; in particular, it calculated the trajectory of the rocket that delivered the pennant of the Soviet Union to the moon. For the M-20 they used automatic address modification, a combination of the arithmetic unit and instruction fetching from memory, and, for the first time, buffer memory was used for the print amounts. The technical performance of the machine was 20 million operations per second.
In 1965 the BESM-4 was built, based on the semiconductor elements, and had software compatibility with the M-20.
In 1967 the first Soviet supercomputer second generation BESM-6 was accepted for mass production. The BESM-6 had a speed of 1 million operations per second.
The BESM-6 was based on the arterial or water supply principle of operation. With its help, the command streams and operands were processed in parallel (up to eight computer operations at various stages). It had associative memory on ultra-fast registers, which reduced the number of calls to the ferrite memory and enabled performing local optimization of the calculations in the dynamics of the account. The BESM-6 also had interleaved RAM, consisting of standalone modules, making it possible to simultaneously access the memory blocks in several directions.
Multi-program modes helped to solve some problems with the given priorities, and made possible the mathematical transformation of an instrumental mechanism into a physical address and to dynamically allocate RAM in a computing operating system. The computer had the typical principle of sheet memory organization, and from that basis they developed the protection mechanisms of the numbers and commands, an advanced system of interrupts with an automatic switch into the solution of another problem, and external devices instructions and the monitoring of its work. They used 60 thousand transistors and 180 thousand semiconductor diodes in the BESM-6 electronic circuits. The BESM-6 hardware components were new at the time and laid the foundation for the computer architecture of the third and fourth generations.
In 1990 a copy of the BESM-6 was installed at the Science Museum of London as a representation of the best supercomputer of its time in Europe.
It’s worth noting that among the specialized computers designed by IPM and CT there were computers to build anti missile defense (AMD): "Diana-1" and "Diana-2" (1955) (head design engineer V.S. Burtsev). Those were the computers of the serial operation with the commute processing program, with automatic data reading from a surveillance radar station, separation of the object from the noise, and with the trajectory calculation of a possible target.
In 1958 M-40 was developed with floating-cycle operations management and an interruptions system. For the first time it had the combination of exchange operations, a multiplex channel of exchange, operation in a closed loop of control as the control link, and operation of the remote objects by microwave-duplex links. Also for the first time, M-40 had equipment for time storage; they also used ferrite-transistor elements and fixed point. Performance of the M-40 was 40 million operations per second. A few years later the M-50 was developed as its modification, which was designed for an experimental AMD system.
Subsequently, in 1963, the 5E92 computer was developed (by Lebedev, V.S.Burtsev etc.), with extensive use of ferrite-transistor elements in the low-frequency devices, and with the use of specially- designed control and recording equipment, with possible remote information recording from the carrier links.
Its modification, 5E92b, was released in 1965, and became one of the first fully semiconductor-based computers. Performance of the large computer was 500 million operations per second, and the speed of the small-sized computer was 37 thousand operations per second. The 5E92b computer formed the base of the Head Commanding Data Center (HCDC) of the Soviet Union’s anti missile defense system.
In 1970 IPM and CT introduced the 5E65 computer – a mobile high-duty computer system for special use, which provided research in real-time in the field with a high degree of reliability because of its use of memory with non-destructive readout, full hardware control and the elimination of the consequences of failure. The effectiveness of the computational process was defined by the variable word length and the pushdown organization of arithmetic unit. Using the system, various studies were carried out of airborne radio navigation and measuring devices in the atmosphere and in space.
In 1973 they presented the new modification: 5E67, a mobile high-duty multicomputer complex with a common field of external memory and hard and software reconfigurations at the machine level. It performed its work in difficult climatic conditions, and also provided unique radio measurements of moving objects in the upper layers of the atmosphere in real time.
The last computer in Sergey A. Lebedev’s lifetime was the 5E26 (Lebedev, V.S.Burtsev, E.A.Krivosheev and so on.), the first mobile high-duty multi-processor structure with modular computer memory, which was approved for series production. It was easily adapted to different performance requirements and featured easily adapted memory in the control systems for special use. It was the first computer with automatic backup at the module level, which provided recovery of computing processes when malfunctions and equipment failures occurred in the control systems. It also worked in real time and was supplied by advanced software, the efficient automation programming system and the ability to work with high-level languages. In 5E26 they implemented nonvolatile memory of commands with the ability to overwrite data records by external electrical equipment, and introduced effective system operation with two-level fault cell isolation, providing efficient device recovery by intermediate staff.
Creating the 5E26 computer was the basis for the development of the super-computer, "Elbrus.” The creation of "Elbrus" completed the USSR’s anti missile defense system. The name for the new computer was proposed by Lebedev, but he was not able to take part in its development.
Sergey Lebedev died on July 3, 1974, in Moscow.
In the year of S.A. Lebedev’s 95th anniversary, his achievements were recognized abroad. As a pioneer of computer science, he was awarded the medal of the International Computer Society with the words: "Sergey Lebedev – 1902-1974. The developer and design engineer of the first computer in the Soviet Union. The founder of the Soviet computer industry."
Sergey A. Lebedev
Sergey Lebedev was the developer of the first fully operational electronic computer in continental Europe with programs stored in memory (MESM), and one of the developers of the world's first digital electronic computers that featured dynamic calculation re-programming. Under his leadership and with his direct participation, 18 different computers were created and 15 of them put into production. The experience of Sergey Lebedev was unique in that it covered the period from the creation of the first thermionic valve computers (which were performing only hundreds or thousands of transactions per second) up to the ultra high-speed super computers on large-scale integrated circuits.