Search Machines

As these examples suggest, the GA is not a universal optimizer, guaranteed to work with any kind of problem. It is true that, even when the problem is poorly suited to a GA approach, a slow drift toward good solutions may occur, because the GA may operate as a slightly intelligent random search machine however, this does not imply that the GA would outperform other algorithms, whatever the problem. 

The search machine complex, which grows at the rate of two computers for every 700,000 substances, has proven to be a solid corner-stone for structure searching the CAS databases over the past seven years. However, the following factors have convinced us to replace it with a new system ... 

The search machine hardware and foundation software is becoming obsolete. 

This paper summarises the principal features of the search machine architecture. It then describes the search engines and how they address the new needs and environment that have emerged over the past several years. 

The CAS search machine architecture was shaped by several factors ... 

The software environment which was used for developing the search machines was one that CAS had developed starting with the single user DEC RT-11 operating system. Two previous CAS project efforts had built a multitasking and memory management capability on top of RT-11 which was called MERT-11 (Memory Extended RT-11). Application code developed under MERT-11 is written in the DEC Macro-11 Assembler language. 

Figure 3 provides an overview of the search machine configuration and is an aid in understanding the discussion in the next several sections. 

The iterative search machines are PDP 11/44 computers with 512K memory. The PDP ll/44s were chosen because they allowed more memory than others in the PDP 11 line and because they had the best price-performance ratio in the PDP 11 family at the time. The memory is needed because the iteractive machines are more heavily involved in the networking, because they must handle potentially very large query connection tables and because they must keep as much of the index to the connection table file as possible in memory to reduce I/O to one or fewer reads per connection table access. 

The disks are 256 MB (formatted) CDC disks with a transfer rate of 1.2MB per second. Each of the iterative search computers is connected to the IBM 3705 communications processor and to its screen search machine with 9600 bps serial links. 

Over the lifetime of the system the response time has remained at a constant five to six minutes per search although usage has continued to increase steadily. The cost of the computing equipment has generally dropped, especially as it has become available on the used equipment market. The architecture has proven to be readily operable. Availability and reliability have been generally comparable to the CAS IBM mainframe environment - currently availability is in the area of 99 per cent for the search machines. Some tuning and operational enhancements were implemented in the early years, but such activity has subsided considerably in recent years. 

A number of things have changed since the search machine was designed and implemented in the late 1970s and early 1980s. 

As a result of all of these factors, the second generation architecture, which is being called the search engine complex , is considerably different from the first generation search machine . 

In 1986 a cost study was conducted to compare the cost of searching the CAS Registry File on a parallel bank of Motorola 68020-based minicomputers versus the cost of searching the file on an IBM mainframe. The cumulative costs over a five year period were noticeably higher on the mainframe if response times comparable to the five minute search machine response times were desired. If response times were reduced from about five minutes to about one minute for the average search, then cumulative mainframe costs were more than twice as great as the minicomputer costs. 

When our key words are known, they become the key to the Internet and its resources. The process may begin with the use of any search tools, for example Google or Google Scholar, or any other specialized search machine or agent. In fact, although the selection of a search machine is important, how we use it is even more important. Here we have several helpful heuristics in our quest to learn about the problem domain ... 

The first assignment in the course is a literature search in Portuguese on Who s Who in Green Chemistry, carried out by individual students. This exercise reveals the disparity of entries on popular search machines and the... 

Conformational Flexibility in 3D Structure Searching Machine Learning Techniques in Chemistry Molecular Dynamics Simulations of Nucleic Acids Nucleic Acid Conformation and Flexibility Modeling Using Molecular... 

---