Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 7 years ago.
Improve this question
Looking at the prices for RDS instances, the multi-AZ instances cost double. Having a production environment in mind, does it worth it?
What uptime should I expect from a single-AZ instance, as opposed to a multi-AZ one? Has anyone had experience running a production DB on both single and multi availability zones?
We have a multi-AZ production deployment with AWS RDS and it's been working fabulously well for the last 3 years.
The multi-AZ catalog page clearly lists out the benefit of using a multi-az vs single RDS deployment. One of the most important aspects of running multi-az is the fact that if one of the AZ in a region goes down, the production application traffic is automatically routed to the RDS in the alternate AZ. Also, DB maintenance and upgrades are applied to the RDS per AZ basis (for a multi-AZ RDS) without impacting uptime.
With respect to cost, it is totally up to the nature of the application as to how much is the degree of downtime tolerance that it can sustain. It's a cost vs uptime trade-off.
Related
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 2 years ago.
Improve this question
I'm trying to determine wether or not to use PostGres or Cassandra data store. Data will be inserted to this new store initially via :
1. A batch process to insert 100MB per day.
2. At a later point the existing batch process + CRUD operations at a rate of 20 operations per second.
From reading articles & various conversations with other I have determined:
**Cassandra**
Type: NoSql
Read Speed: Fast
Write Speed: Slow
Storage: Distributed Cluster
Scaling: Horizontal
**PostGres**
Type: Relational
Storage: Single Instance
Scaling: Horizontal
Some resources I have been reading :
https://www.quora.com/How-do-you-compare-Postgres-to-Cassandra-for-big-data-application
https://www.quora.com/How-do-you-approach-choosing-a-database-architecture-and-design
https://www.thegeekstuff.com/2014/01/sql-vs-nosql-db/?utm_source=tuicool
What other considerations should be taken before making the decision ? Are there other data points I should include in the decision process such as determining the expected number reads, writes, updates, deletes from the table ?
I could utilize PostGres and then migrate to Cassandra at some other point but would prefer to avoid the overhead of a DB migration process.
I work with postgres, because it is more customizable for developers and it is a nosql database combined with the most beautiful parts with sql db.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 3 years ago.
Improve this question
I am trying to implement sql server database replication between 2 branch servers to a Head Office Server.
My application is a distributed one, the main application is hosted on head office which is controlling the masters and final approvals. The branch servers are located on two other countries, which are used to enter daily transactions.
Since internet bandwidth is too slow, I am planning to run the replication only on off hours (ie.. night 12 AM to Morning 8 AM). During business hours it is difficult to synchronize. All tables are designed such a way to validate and avoid duplication or other errors.
Also there are chances of internet outage for couple of days.. may be up to a week.
There are three type of tables,
BI Directional - Needs to sync between both sides(HO to branch and
branch to HO, Approvals)
Sync from Branch to HO (Transactions)
Sync from HO to Branches (Masters)
When I configure replication, I am confused between different types of replication such as snapshot, transactional and merge replication.
Can anybody suggest which one is the best method for my model
I am also facing some issues with Primary key and foreign keys lost after configuring replication.. Any idea why this is happened..?
Transaction replication is best for one way sync and Merge for Bi directional sync, will be the best options.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 4 years ago.
Improve this question
Does it make sense to change from MongoDB to SQL Server in the MEAN full stack (Node Express and Angular)?
In today's data environment, the unfortunate truth is that there is no one "right" data appliance for all use cases. At my last job, we used Hadoop for our streaming data, Couchbase as a web data caching layer, Hive when we needed the illusion of a relational database, SQL Server on Azure when we needed a temporal data store with more than the illusion of relational characteristics, on prem SQL Servers when for the persistence layer where we needed actual relational data, and a Netezza cube for reporting. There's not a handy acronym for that hot mess.
I will say that a lot of shops that are primarily NoSQL focused like Postgres as an RDBMS addition since it keeps with the whole open source philosophy, but your mileage may vary.
The TLDR; answer to your question is that it depends on your use case(s), and there's no avoiding painful hours of research and probably a few false starts before you land on a solution that works for you.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
This question does not appear to be about a specific programming problem, a software algorithm, or software tools primarily used by programmers. If you believe the question would be on-topic on another Stack Exchange site, you can leave a comment to explain where the question may be able to be answered.
Closed 7 years ago.
Improve this question
I have an AWS RDS SQL Server RDS instance running with mirroring (Multi-AZ). Typically the mirror failover server has to be declared in the connection string as per http://msdn.microsoft.com/en-us/library/5h52hef8(v=vs.110).aspx
I can't find what the "Failover Partner" should be in the aws console.
Does aws/rds take care of this behind the scenes so i only have to declare 1 server?
This should be handled by AWS for you using DNS changes only:
The primary and standby instances use the same endpoint whose physical
network address transitions to the mirror as part of the failover
process, so you do not have to reconfigure your application or set up
multiple endpoints when a failover occurs.
From here:
http://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/CHAP_SQLServer.html#SQLServer.Concepts.General.Mirroring
Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 9 years ago.
Improve this question
Quick Question: Is it possible to mirror a database without downtime?
Long Question: I have a database on production being used by quite a few clients. The previous management did not implemented any kind of redundancy or high-availability strategy (no clustering... not even storage!!!), and now, as business grows, this is becoming a huge liability... as a emergency act I'm considering mirroring the database... The main problem is that I cannot take down the database. That would imply on some legal/financial problems due to some previous SLA agreements... So, can I mirror a database without taking it down?
Extra info:
The SQL Server version is 2008 R2 Enterprise.
The instance consist of one database only (it's a multi-tenant database)
The database infrastructure consist of one physical server running windows 2008 R2 (standalone server). It's not a cluster nor a VM and theres no storage behind it... all data is inside it's only 2TB disk...
The Database size (.mdf) is about 170 GB...
There's about 100 transactions each second
There's no hours when usage goes down... business are 24/7...
Yes, this totally looks like that environment a developer would create on their machine...
This is all in books online. Hopefully you can try in development, or on your desktop first. There are quite a few things to consider, so you should give it a thorough read.
Yes, you can mirror without taking it down. You restore a backup to the mirror and use NORECOVERY option. You then setup the mirroring and transactions start to move from primary to mirror. You'll need to set up any other things needed on that remote server - logins, jobs, etc.
Transfer logins using this method, so you get matching SIDs between servers.