![]() | |
![]() |
| | Thread Tools | Display Modes |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
Hello, After having quite some research, I found out that Active/Active servers are not recommended by most of people as clustering is mainly is a failover technology. However, there was an opinion of using Clustering for load balance in a single instance SQL 2005 database. Based on this, I had the though that I should go for an Active/Active/Passive model of 3 nodes, 2 for load balancing, (faster querys) and one for failover. My research has shown that the failover works if the most half the servers fail, and I think this can cover the issue. I am quite worried though about load balancing. I believe that clustering would not help, and another kind of technology, or even better servers would do the trick. Another thing that makes me think hardly about the load balancing, is the transaction log. I believe that a transaction log for an SQL instance, should be tampered only by one server. Would an Active/Active server be different? Searching into the internet, I found out that there are lots of information and quite vague about this subject, and I would really need some help on this, to see if I should go on implementing such a solution. Being new in that stuff, I would appreciate a solution that would give optimum speed (through clustering or not) and some kind of reliability. Thanks a lot! |
#3
| |||
| |||
|
|
Hello, After having quite some research, I found out that Active/Active servers are not recommended by most of people as clustering is mainly is a failover technology. However, there was an opinion of using Clustering for load balance in a single instance SQL 2005 database. Based on this, I had the though that I should go for an Active/Active/Passive model of 3 nodes, 2 for load balancing, (faster querys) and one for failover. My research has shown that the failover works if the most half the servers fail, and I think this can cover the issue. I am quite worried though about load balancing. I believe that clustering would not help, and another kind of technology, or even better servers would do the trick. Another thing that makes me think hardly about the load balancing, is the transaction log. I believe that a transaction log for an SQL instance, should be tampered only by one server. Would an Active/Active server be different? Searching into the internet, I found out that there are lots of information and quite vague about this subject, and I would really need some help on this, to see if I should go on implementing such a solution. Being new in that stuff, I would appreciate a solution that would give optimum speed (through clustering or not) and some kind of reliability. Thanks a lot! |
#4
| |||
| |||
|
|
Hello, After having quite some research, I found out that Active/Active servers are not recommended by most of people as clustering is mainly is a failover technology. However, there was an opinion of using Clustering for load balance in a single instance SQL 2005 database. Based on this, I had the though that I should go for an Active/Active/Passive model of 3 nodes, 2 for load balancing, (faster querys) and one for failover. My research has shown that the failover works if the most half the servers fail, and I think this can cover the issue. I am quite worried though about load balancing. I believe that clustering would not help, and another kind of technology, or even better servers would do the trick. Another thing that makes me think hardly about the load balancing, is the transaction log. I believe that a transaction log for an SQL instance, should be tampered only by one server. Would an Active/Active server be different? Searching into the internet, I found out that there are lots of information and quite vague about this subject, and I would really need some help on this, to see if I should go on implementing such a solution. Being new in that stuff, I would appreciate a solution that would give optimum speed (through clustering or not) and some kind of reliability. Thanks a lot! |
#5
| |||
| |||
|
|
Hello, After having quite some research, I found out that Active/Active servers are not recommended by most of people as clustering is mainly is a failover technology. However, there was an opinion of using Clustering for load balance in a single instance SQL 2005 database. Based on this, I had the though that I should go for an Active/Active/Passive model of 3 nodes, 2 for load balancing, (faster querys) and one for failover. My research has shown that the failover works if the most half the servers fail, and I think this can cover the issue. I am quite worried though about load balancing. I believe that clustering would not help, and another kind of technology, or even better servers would do the trick. Another thing that makes me think hardly about the load balancing, is the transaction log. I believe that a transaction log for an SQL instance, should be tampered only by one server. Would an Active/Active server be different? Searching into the internet, I found out that there are lots of information and quite vague about this subject, and I would really need some help on this, to see if I should go on implementing such a solution. Being new in that stuff, I would appreciate a solution that would give optimum speed (through clustering or not) and some kind of reliability. Thanks a lot! |
#6
| |||
| |||
|
|
Hello, After having quite some research, I found out that Active/Active servers are not recommended by most of people as clustering is mainly is a failover technology. However, there was an opinion of using Clustering for load balance in a single instance SQL 2005 database. Based on this, I had the though that I should go for an Active/Active/Passive model of 3 nodes, 2 for load balancing, (faster querys) and one for failover. My research has shown that the failover works if the most half the servers fail, and I think this can cover the issue. I am quite worried though about load balancing. I believe that clustering would not help, and another kind of technology, or even better servers would do the trick. Another thing that makes me think hardly about the load balancing, is the transaction log. I believe that a transaction log for an SQL instance, should be tampered only by one server. Would an Active/Active server be different? Searching into the internet, I found out that there are lots of information and quite vague about this subject, and I would really need some help on this, to see if I should go on implementing such a solution. Being new in that stuff, I would appreciate a solution that would give optimum speed (through clustering or not) and some kind of reliability. Thanks a lot! |
#7
| |||
| |||
|
|
Hello, After having quite some research, I found out that Active/Active servers are not recommended by most of people as clustering is mainly is a failover technology. However, there was an opinion of using Clustering for load balance in a single instance SQL 2005 database. Based on this, I had the though that I should go for an Active/Active/Passive model of 3 nodes, 2 for load balancing, (faster querys) and one for failover. My research has shown that the failover works if the most half the servers fail, and I think this can cover the issue. I am quite worried though about load balancing. I believe that clustering would not help, and another kind of technology, or even better servers would do the trick. Another thing that makes me think hardly about the load balancing, is the transaction log. I believe that a transaction log for an SQL instance, should be tampered only by one server. Would an Active/Active server be different? Searching into the internet, I found out that there are lots of information and quite vague about this subject, and I would really need some help on this, to see if I should go on implementing such a solution. Being new in that stuff, I would appreciate a solution that would give optimum speed (through clustering or not) and some kind of reliability. Thanks a lot! |
#8
| |||
| |||
|
|
Hello, After having quite some research, I found out that Active/Active servers are not recommended by most of people as clustering is mainly is a failover technology. However, there was an opinion of using Clustering for load balance in a single instance SQL 2005 database. Based on this, I had the though that I should go for an Active/Active/Passive model of 3 nodes, 2 for load balancing, (faster querys) and one for failover. My research has shown that the failover works if the most half the servers fail, and I think this can cover the issue. I am quite worried though about load balancing. I believe that clustering would not help, and another kind of technology, or even better servers would do the trick. Another thing that makes me think hardly about the load balancing, is the transaction log. I believe that a transaction log for an SQL instance, should be tampered only by one server. Would an Active/Active server be different? Searching into the internet, I found out that there are lots of information and quite vague about this subject, and I would really need some help on this, to see if I should go on implementing such a solution. Being new in that stuff, I would appreciate a solution that would give optimum speed (through clustering or not) and some kind of reliability. Thanks a lot! |
#9
| |||
| |||
|
|
Hello, After having quite some research, I found out that Active/Active servers are not recommended by most of people as clustering is mainly is a failover technology. However, there was an opinion of using Clustering for load balance in a single instance SQL 2005 database. Based on this, I had the though that I should go for an Active/Active/Passive model of 3 nodes, 2 for load balancing, (faster querys) and one for failover. My research has shown that the failover works if the most half the servers fail, and I think this can cover the issue. I am quite worried though about load balancing. I believe that clustering would not help, and another kind of technology, or even better servers would do the trick. Another thing that makes me think hardly about the load balancing, is the transaction log. I believe that a transaction log for an SQL instance, should be tampered only by one server. Would an Active/Active server be different? Searching into the internet, I found out that there are lots of information and quite vague about this subject, and I would really need some help on this, to see if I should go on implementing such a solution. Being new in that stuff, I would appreciate a solution that would give optimum speed (through clustering or not) and some kind of reliability. Thanks a lot! |
#10
| |||
| |||
|
|
Hello, After having quite some research, I found out that Active/Active servers are not recommended by most of people as clustering is mainly is a failover technology. However, there was an opinion of using Clustering for load balance in a single instance SQL 2005 database. Based on this, I had the though that I should go for an Active/Active/Passive model of 3 nodes, 2 for load balancing, (faster querys) and one for failover. My research has shown that the failover works if the most half the servers fail, and I think this can cover the issue. I am quite worried though about load balancing. I believe that clustering would not help, and another kind of technology, or even better servers would do the trick. Another thing that makes me think hardly about the load balancing, is the transaction log. I believe that a transaction log for an SQL instance, should be tampered only by one server. Would an Active/Active server be different? Searching into the internet, I found out that there are lots of information and quite vague about this subject, and I would really need some help on this, to see if I should go on implementing such a solution. Being new in that stuff, I would appreciate a solution that would give optimum speed (through clustering or not) and some kind of reliability. Thanks a lot! |
![]() |
| Thread Tools | |
| Display Modes | |
| |