【Weekly Question 20090511】Please list the most prominent features (or advantages) of each DCS.
Thread Content
This post was last edited by wopale3 on 2009-5-15 at 11:56. In order to establish a strong foothold in the market, DCS manufacturers have introduced many additional functions – some of these involve complex operations, while others add various communication interfaces, and so on. End-users end up paying a high price in terms of money and effort. What’s worse, these additions can lead to high system loads and instability. Therefore, not all upgrades are successful! Please share your experiences based on actual situations, using the following format: 1. Name of the system manufacturer; 2. Time when the issue occurred; 3. Version being used; 4. Advantages brought about; 5. Disadvantages incurred; 6. Other details... We hope everyone will express their opinions freely, actively participate in the discussion, and those who offer valuable insights will be rewarded generously. Encourage originality
Let me start with: 1. Emerson; 2. May 2006; 3. DELTAV6.3.2 version; 4. The configuration is extremely convenient and easy to use; 5. There was one instance of a network disruption caused by a timing mismatch.
It seems that I haven’t encountered any useless features for now; rather, I feel that the functions available are too weak. There are many things that are actually quite useful, but the developers didn’t think of them, which makes it very difficult for the debugging engineers. . . So far, no redundant functions in that DCS have been identified; it’s just that everything isn’t done well enough.
How about rephrasing it? Everyone, please list the most prominent features (or advantages) of each DCS. I think everyone can discuss this then; for example, Honeywell’s solutions, Profiloop’s products, and ABB’s AS servers with triple fault tolerance in a 1:1:1 configuration. Emerson’s AMS . FF – CAD configuration of Emerson-WestingHouse. Siemens’ Profibus PA ring. . . .
The last edit to this post was made by qiaopeng1231 on 2009-5-15 at 20:21. Let’s start with Siemens PCS7. PCS7’s WebServer supports simple load balancing. Assume that a Web Server is a 4-way cluster. There are already 44 IE clients connected. Assume Server-A has 10 connections with a load rate of 20% ； Server-B has 11 connections, with a load rate of 22% ； Server-C has 12 connections, with a load rate of 25% ； Server-D has 11 connections, with a compliance rate of 21%. Now, new IE clients need to connect as well. The system will automatically determine the load rate of each server. New connections are prioritized to be assigned to the server with the lowest load rate. This is a typical load balancing system that gives priority to low load rates. There is a new IE connection request, which will connect to Server-A; then another IE connection request comes in, and it will connect to Server-D. And so on. This ensures that the load on each server is roughly balanced. In fact, many of the large websites we visit follow this pattern. Although there is only one website address, there are numerous server clusters behind it. Load balancing is achieved among the members of the cluster.
This post was last edited by qiaopeng1231 on 2009-5-16 21:26: Honeywell FTE. Suppose there is CPU-A, whose Ethernet module is FTEM-A (with Port1 and Port2); and CPU-B, whose Ethernet module is FTEM-B (with Port1 and Port2). There are switches Switch1 and Switch2. There is PC-001 (with network cards PC-001-1, PC-001-2), and PC-002 (with network cards PC-002-1, PC-002-2). FTEM-A-Port1 / FTEM-B-Port1 ---》 Switch1 ---》PC-001-1/PC-002-1 FTEM-A-Port2 / FTEM-B-Port2 ---》 Switch2 ---》PC-001-2/PC-002-2 Regarding the IP issue. . . The two network cards of PC-001 and PC-002 share one IP address; they behave as if they were a single network card, remaining transparent to external entities, allowing either network card to be damaged. Switch 1 and Switch2 are also redundant to each other, allowing either switch to be damaged. Port1 and Port2 of FTEM-A are fault-tolerant to each other, functioning like a single port. It is an IP. For applications and communications, just look for 1 IP address. Regardless of which specific port it is. Port1 and Port2 of FTEM-B are fault-tolerant to each other, functioning like a single port. It is an IP. For applications and communications, just look for 1 IP address. Regardless of which specific port it is. However, the applications and communication nodes communicate only with the FTEM of the main CPU at all times. . . The IPs of FTEM-A and FTEM-B can also be considered as 1 IP. Assuming CPU-A is the primary one, the IP of FTEM-A is N, and the IP of FTEM-B is N+1. If the CPU switches and CPU-B becomes the primary one, then the IP of FTEM-A is N+1, and the IP of FTEM-B is N. At any time, the communication node should simply go and communicate with the FTEM of the CPU with IP address N. In this way, the redundancy of the CPU is also transparent, just as if communicating with a single CPU. 2 FTE modules, 4 links; from the outside it appears to be just 1 unit, with all switching and fault tolerance functions being handled internally by the system itself. External communication access has become extremely simple.