Share this
How To Build Scalable Database Architectures
by Shakir Sadikali on Mar 12, 2008 12:00:00 AM
I’ve found lately that munching on carrots with French dressing is more satisfying than broccoli. Maybe it’s the tang-and-crunch combination. In any case, I was crunching away yesterday while thinking about how to answer a question one of our newer start-up clients asked me.
No one has ever come out and formally asked me for a document that states “Best Practices to Scale Application X”. It is an unusual demand, since it’s something many of us at Pythian have implemented, but it’s been more of an ad hoc, iterative process — and rightly so, since architectures must be so organic, and so tailored to the application. What’s more, no one has ever brought us on board so early in the game that we have a hand in actually — gasp! — doing the design and data-model from the get-go. Woo hoo!
Now, a little background. I have built and maintained a few systems. Some of them even supported over 100k concurrent users. These databases didn’t run RAC either (although I do support two very high profile RAC environments now). So, having been in the trenches and knowing what it takes to make a DB move, I got to thinking about some of the basic fundamentals. There are always rules of thumb, right? This is what you need to know to start with building a scalable high-performance system based on stuff that I’ve seen. Obviously, this assumes a database-centric app. Let’s start with the first ten principles.
1. Bind your variables.
Hard parsing your queries uses obscene amounts of CPU. Don’t believe me, just Ask Tom.
I was asked to audit a client application running on a five-node cluster. They had five nodes because the application was very CPU-intensive (which was good, because the reason to “go RAC” should be either to achieve redundancy, or to scale-out your CPU power).
Our audit uncovered huge amounts of waste. All the queries were unbound. Working on their app to bind all the variables (maybe two weeks of work), we were able to get us down to a single non-RAC database. How much was lazy coding costing this client?
What’s worse, if you use SGA_TARGET
, not binding will kill your performance more than it’s already being killed by the hard parses. Your SGA gets bloated because all your SQL isn’t getting re-used, and as a result, your DB cache shrinks. It’s adding insult to injury.
2. Don’t run reports off your live tables. EVER.
This is a big one for many start-ups. I see it all the time. The have nightly reports that run and grab stuff like, “Here’s a breakdown of the membership on my site by geographic region,” or something. “Great, I’ll just run select count(*), country from member group by country;
.” This kind of thing works great when you have 100 users. Now that you have 10 million, it doesn’t work so great. This is a full tablescan (usually) and it holds read consistency. This type of query will bring your site down.
Get a second server, or another DB on the same hardware, push the data there, and run your reports off it. You can run as many massive queries as you want there, but don’t touch the OLTP part of the app with them. If you need to get the results back to the live site, push them there using PL/SQL or whatever method you fancy.
3. Don’t go crazy doing live updates of things that aren’t needed live.
Most people don’t care to see their last login date that happened three minutes ago. If you have 100k concurrent users and you’re updating columns like this (that are actually useful to see the stickiness of your site month to month) don’t do it live on login. Make it asynchronous using batch jobs.
4. If your site is read-heavy vs. write-heavy, design it appropriately with this in mind.
Do your best to avoid mixing the two types of operations. You can write on a master and read from remote sites. Oracle Advanced Replication is great for this.
5. Use connection pooling.
I/O kills a system. If you have 100k people all logging on and hitting your systems, you can’t expect to spawn a new connection for each of them. In a nutshell, each time a connection is created, Oracle has to stat all the datafiles for that spawned process. This I/O is huge, and doing it 100k times will kill you. Find a way to connection pool.
6. Write smart SQL.
Look at your explain plan and the resources your query is using. The foundation of all query performance is the buffer_get
. It doesn’t matter whether it’s a disk buffer_get
or a memory buffer_get
. If you get your query to request the fewest buffer_get
s possible, the overall performance impact will be beneficial. It all starts with the buffer_get
.
Make sure you’re indexed properly, don’t use bitmaps on volatile data, use compressed indexes where appropriate, etc. Know the difference between a hash join and nested loop. This one is HUGE. You know your data better than anyone. You need to know how much data you’re grabbing from individual tables, and this should dictate what type of table access to use. Hash joins are extremely useful for very large queries.
7. Use PL/SQL.
The database is hugely powerful. Let it do some of the business logic using PL/SQL. It’s built for it.
8. Set HugePages.
If you’re running Linux, set HugePages and free up your RAM to be usable real work, instead of wasting it on addressing. Christo Kutrovsky made an excellent presentation on this topic.
9. Don’t run batch processes off live tables.
When you have tables that are actively being written into, but which you also need to process to get the results out, don’t open a cursor, process it, and then commit three minutes later. You’re holding the application up for those minutes. Create a view on the table. When you need to process, flip the view to another identical table so that the inserts can keep coming in. You’re now free to process the original table at your leisure.
10. Get the data-model right (or at least, not blatantly bad).
This is the guts of your app. Avoid circular references. Unfortunately, many constraints may need to be enforced in the application. Don’t be cheap with your columns. Don’t try to make one column do too many things. Don’t over-normalize. You get speed from having redundancy because you can avoid joins. Index properly and remember in which direction your relationships go. (I think I’ll need to make a blog post — or a few — about just this topic because it’s so important.)
I realize I’m generalizing a little. Okay, a lot. I’ll post again with a little more detail on some of these with examples of good/not-so-good code. In the meantime, I’d love to hear about tricks you have used. I myself have come across so many strange things, I’ll see if I can dig up a few horrors.
Cheers!
Share this
- Technical Track (967)
- Oracle (412)
- MySQL (140)
- Cloud (128)
- Microsoft SQL Server (117)
- Open Source (90)
- Google Cloud (81)
- Microsoft Azure (63)
- Amazon Web Services (AWS) (58)
- Big Data (52)
- Google Cloud Platform (46)
- Cassandra (44)
- DevOps (41)
- Pythian (33)
- Linux (30)
- Database (26)
- Performance (25)
- Podcasts (25)
- Site Reliability Engineering (25)
- PostgreSQL (24)
- Oracle E-Business Suite (23)
- Oracle Database (22)
- Docker (21)
- DBA (20)
- Security (20)
- Exadata (18)
- MongoDB (18)
- Oracle Cloud Infrastructure (OCI) (18)
- Oracle Exadata (18)
- Automation (17)
- Hadoop (16)
- Oracleebs (16)
- Amazon RDS (15)
- Ansible (15)
- Snowflake (15)
- ASM (13)
- Artificial Intelligence (AI) (13)
- BigQuery (13)
- Replication (13)
- Advanced Analytics (12)
- Data (12)
- GenAI (12)
- Kubernetes (12)
- LLM (12)
- Authentication, SSO and MFA (11)
- Cloud Migration (11)
- Machine Learning (11)
- Rman (11)
- Datascape Podcast (10)
- Monitoring (10)
- Oracle Applications (10)
- Apache Cassandra (9)
- ChatGPT (9)
- Data Guard (9)
- Infrastructure (9)
- Python (9)
- Series (9)
- AWR (8)
- High Availability (8)
- Oracle EBS (8)
- Oracle Enterprise Manager (OEM) (8)
- Percona (8)
- Apache Beam (7)
- Data Governance (7)
- Innodb (7)
- Microsoft Azure SQL Database (7)
- Migration (7)
- Myrocks (7)
- Performance Tuning (7)
- Data Enablement (6)
- Data Visualization (6)
- Database Performance (6)
- Oracle Enterprise Manager (6)
- Orchestrator (6)
- RocksDB (6)
- Serverless (6)
- Azure Data Factory (5)
- Azure Synapse Analytics (5)
- Covid-19 (5)
- Disaster Recovery (5)
- Generative AI (5)
- Google BigQuery (5)
- Mariadb (5)
- Microsoft (5)
- Scala (5)
- Windows (5)
- Xtrabackup (5)
- Airflow (4)
- Analytics (4)
- Apex (4)
- Cloud Security (4)
- Cloud Spanner (4)
- CockroachDB (4)
- Data Management (4)
- Data Pipeline (4)
- Data Security (4)
- Data Strategy (4)
- Database Administrator (4)
- Database Management (4)
- Database Migration (4)
- Dataflow (4)
- Fusion Middleware (4)
- Google (4)
- Oracle Autonomous Database (Adb) (4)
- Oracle Cloud (4)
- Prometheus (4)
- Redhat (4)
- Slob (4)
- Ssl (4)
- Terraform (4)
- Amazon Relational Database Service (Rds) (3)
- Apache Kafka (3)
- Apexexport (3)
- Aurora (3)
- Business Intelligence (3)
- Cloud Armor (3)
- Cloud Database (3)
- Cloud FinOps (3)
- Cosmos Db (3)
- Data Analytics (3)
- Data Integration (3)
- Database Monitoring (3)
- Database Troubleshooting (3)
- Database Upgrade (3)
- Databases (3)
- Dataops (3)
- Digital Transformation (3)
- ERP (3)
- Google Chrome (3)
- Google Cloud Sql (3)
- Google Workspace (3)
- Graphite (3)
- Heterogeneous Database Migration (3)
- Liquibase (3)
- Oracle Data Guard (3)
- Oracle Live Sql (3)
- Oracle Rac (3)
- Perl (3)
- Rdbms (3)
- Remote Teams (3)
- S3 (3)
- SAP (3)
- Tensorflow (3)
- Adf (2)
- Adop (2)
- Amazon Data Migration Service (2)
- Amazon Ec2 (2)
- Amazon S3 (2)
- Apache Flink (2)
- Ashdump (2)
- Atp (2)
- Autonomous (2)
- Awr Data Mining (2)
- Cloud Cost Optimization (2)
- Cloud Data Fusion (2)
- Cloud Hosting (2)
- Cloud Infrastructure (2)
- Cloud Shell (2)
- Cloud Sql (2)
- Conferences (2)
- Cosmosdb (2)
- Cost Management (2)
- Cyber Security (2)
- Data Analysis (2)
- Data Discovery (2)
- Data Engineering (2)
- Data Migration (2)
- Data Modeling (2)
- Data Quality (2)
- Data Streaming (2)
- Data Warehouse (2)
- Database Consulting (2)
- Database Migrations (2)
- Dataguard (2)
- Docker-Composer (2)
- Enterprise Data Platform (EDP) (2)
- Etl (2)
- Events (2)
- Gemini (2)
- Health Check (2)
- Infrastructure As Code (2)
- Innodb Cluster (2)
- Innodb File Structure (2)
- Innodb Group Replication (2)
- NLP (2)
- Neo4J (2)
- Nosql (2)
- Open Source Database (2)
- Oracle Datase (2)
- Oracle Extended Manager (Oem) (2)
- Oracle Flashback (2)
- Oracle Forms (2)
- Oracle Installation (2)
- Oracle Io Testing (2)
- Podcast (2)
- Power Bi (2)
- Redshift (2)
- Remote DBA (2)
- Remote Sre (2)
- SAP HANA Cloud (2)
- Single Sign-On (2)
- Webinars (2)
- X5 (2)
- Actifio (1)
- Adf Custom Email (1)
- Adrci (1)
- Advanced Data Services (1)
- Afd (1)
- Ahf (1)
- Alloydb (1)
- Amazon (1)
- Amazon Athena (1)
- Amazon Aurora Backtrack (1)
- Amazon Efs (1)
- Amazon Redshift (1)
- Amazon Sagemaker (1)
- Amazon Vpc Flow Logs (1)
- Analysis (1)
- Analytical Models (1)
- Anisble (1)
- Anthos (1)
- Apache (1)
- Apache Nifi (1)
- Apache Spark (1)
- Application Migration (1)
- Ash (1)
- Asmlib (1)
- Atlas CLI (1)
- Awr Mining (1)
- Aws Lake Formation (1)
- Azure Data Lake (1)
- Azure Data Lake Analytics (1)
- Azure Data Lake Store (1)
- Azure Data Migration Service (1)
- Azure OpenAI (1)
- Azure Sql Data Warehouse (1)
- Batches In Cassandra (1)
- Business Insights (1)
- Chown (1)
- Chrome Security (1)
- Cloud Browser (1)
- Cloud Build (1)
- Cloud Consulting (1)
- Cloud Data Warehouse (1)
- Cloud Database Management (1)
- Cloud Dataproc (1)
- Cloud Foundry (1)
- Cloud Manager (1)
- Cloud Networking (1)
- Cloud SQL Replica (1)
- Cloud Scheduler (1)
- Cloud Services (1)
- Cloud Strategies (1)
- Compliance (1)
- Conversational AI (1)
- DAX (1)
- Data Analytics Platform (1)
- Data Box (1)
- Data Classification (1)
- Data Cleansing (1)
- Data Encryption (1)
- Data Estate (1)
- Data Flow Management (1)
- Data Insights (1)
- Data Integrity (1)
- Data Lake (1)
- Data Leader (1)
- Data Lifecycle Management (1)
- Data Lineage (1)
- Data Masking (1)
- Data Mesh (1)
- Data Migration Assistant (1)
- Data Migration Service (1)
- Data Mining (1)
- Data Monetization (1)
- Data Policy (1)
- Data Profiling (1)
- Data Protection (1)
- Data Retention (1)
- Data Safe (1)
- Data Sheets (1)
- Data Summit (1)
- Data Vault (1)
- Data Warehouse Modernization (1)
- Database Auditing (1)
- Database Consultant (1)
- Database Link (1)
- Database Modernization (1)
- Database Provisioning (1)
- Database Provisioning Failed (1)
- Database Replication (1)
- Database Scaling (1)
- Database Schemas (1)
- Database Security (1)
- Databricks (1)
- Datascape 59 (1)
- DeepSeek (1)
- Duet AI (1)
- Edp (1)
- Gcp Compute (1)
- Gcp-Spanner (1)
- Global Analytics (1)
- Google Analytics (1)
- Google Cloud Architecture Framework (1)
- Google Cloud Data Services (1)
- Google Cloud Partner (1)
- Google Cloud Spanner (1)
- Google Cloud VMware Engine (1)
- Google Compute Engine (1)
- Google Dataflow (1)
- Google Datalab (1)
- Google Grab And Go (1)
- Graph Algorithms (1)
- Graph Databases (1)
- Graph Inferences (1)
- Graph Theory (1)
- GraphQL (1)
- Healthcheck (1)
- Information (1)
- Infrastructure As A Code (1)
- Innobackupex (1)
- Innodb Concurrency (1)
- Innodb Flush Method (1)
- It Industry (1)
- Kubeflow (1)
- LMSYS Chatbot Arena (1)
- Linux Host Monitoring (1)
- Linux Storage Appliance (1)
- Looker (1)
- MMLU (1)
- Managed Services (1)
- Migrate (1)
- Migrating Ssis Catalog (1)
- Migration Checklist (1)
- MongoDB Atlas (1)
- MongoDB Compass (1)
- Newsroom (1)
- Nifi (1)
- OPEX (1)
- ORAPKI (1)
- Odbcs (1)
- Odbs (1)
- On-Premises (1)
- Ora-01852 (1)
- Ora-7445 (1)
- Oracle Cursor (1)
- Oracle Database Appliance (1)
- Oracle Database Se2 (1)
- Oracle Database Standard Edition 2 (1)
- Oracle Database Upgrade (1)
- Oracle Database@Google Cloud (1)
- Oracle Exadata Smart Scan (1)
- Oracle Licensing (1)
- Oracle Linux Virtualization Manager (1)
- Oracle Oda (1)
- Oracle Openworld (1)
- Oracle Parallelism (1)
- Oracle RMAN (1)
- Oracle Rdbms (1)
- Oracle Real Application Clusters (1)
- Oracle Reports (1)
- Oracle Security (1)
- Oracle Wallet (1)
- Perfomrance (1)
- Performance Schema (1)
- Policy (1)
- Prompt Engineering (1)
- Public Cloud (1)
- Pythian News (1)
- Rdb (1)
- Replication Compatibility (1)
- Replication Error (1)
- Retail (1)
- Scaling Ir (1)
- Securing Sql Server (1)
- Security Compliance (1)
- Serverless Computing (1)
- Sso (1)
- Tenserflow (1)
- Teradata (1)
- Vertex AI (1)
- Vertica (1)
- Videos (1)
- Workspace Security (1)
- Xbstream (1)
- May 2025 (3)
- March 2025 (2)
- February 2025 (1)
- January 2025 (2)
- December 2024 (1)
- October 2024 (2)
- September 2024 (7)
- August 2024 (4)
- July 2024 (2)
- June 2024 (6)
- May 2024 (3)
- April 2024 (2)
- February 2024 (1)
- January 2024 (11)
- December 2023 (10)
- November 2023 (11)
- October 2023 (10)
- September 2023 (8)
- August 2023 (6)
- July 2023 (2)
- June 2023 (13)
- May 2023 (4)
- April 2023 (6)
- March 2023 (10)
- February 2023 (6)
- January 2023 (5)
- December 2022 (10)
- November 2022 (10)
- October 2022 (10)
- September 2022 (13)
- August 2022 (16)
- July 2022 (12)
- June 2022 (13)
- May 2022 (11)
- April 2022 (4)
- March 2022 (5)
- February 2022 (4)
- January 2022 (14)
- December 2021 (16)
- November 2021 (11)
- October 2021 (6)
- September 2021 (11)
- August 2021 (6)
- July 2021 (9)
- June 2021 (4)
- May 2021 (8)
- April 2021 (16)
- March 2021 (16)
- February 2021 (6)
- January 2021 (12)
- December 2020 (12)
- November 2020 (17)
- October 2020 (11)
- September 2020 (10)
- August 2020 (11)
- July 2020 (13)
- June 2020 (6)
- May 2020 (9)
- April 2020 (18)
- March 2020 (21)
- February 2020 (13)
- January 2020 (15)
- December 2019 (10)
- November 2019 (11)
- October 2019 (12)
- September 2019 (16)
- August 2019 (15)
- July 2019 (10)
- June 2019 (16)
- May 2019 (20)
- April 2019 (21)
- March 2019 (14)
- February 2019 (18)
- January 2019 (18)
- December 2018 (5)
- November 2018 (16)
- October 2018 (12)
- September 2018 (20)
- August 2018 (27)
- July 2018 (31)
- June 2018 (34)
- May 2018 (28)
- April 2018 (27)
- March 2018 (17)
- February 2018 (8)
- January 2018 (20)
- December 2017 (14)
- November 2017 (4)
- October 2017 (1)
- September 2017 (3)
- August 2017 (5)
- July 2017 (4)
- June 2017 (2)
- May 2017 (7)
- April 2017 (7)
- March 2017 (8)
- February 2017 (8)
- January 2017 (5)
- December 2016 (3)
- November 2016 (4)
- October 2016 (8)
- September 2016 (9)
- August 2016 (10)
- July 2016 (9)
- June 2016 (8)
- May 2016 (13)
- April 2016 (16)
- March 2016 (13)
- February 2016 (11)
- January 2016 (6)
- December 2015 (11)
- November 2015 (11)
- October 2015 (5)
- September 2015 (16)
- August 2015 (4)
- July 2015 (1)
- June 2015 (3)
- May 2015 (6)
- April 2015 (5)
- March 2015 (5)
- February 2015 (4)
- January 2015 (3)
- December 2014 (7)
- October 2014 (4)
- September 2014 (6)
- August 2014 (6)
- July 2014 (16)
- June 2014 (7)
- May 2014 (6)
- April 2014 (5)
- March 2014 (4)
- February 2014 (10)
- January 2014 (6)
- December 2013 (8)
- November 2013 (12)
- October 2013 (9)
- September 2013 (6)
- August 2013 (7)
- July 2013 (9)
- June 2013 (7)
- May 2013 (7)
- April 2013 (4)
- March 2013 (7)
- February 2013 (4)
- January 2013 (4)
- December 2012 (6)
- November 2012 (8)
- October 2012 (9)
- September 2012 (3)
- August 2012 (5)
- July 2012 (5)
- June 2012 (7)
- May 2012 (11)
- April 2012 (1)
- March 2012 (8)
- February 2012 (1)
- January 2012 (6)
- December 2011 (8)
- November 2011 (5)
- October 2011 (9)
- September 2011 (6)
- August 2011 (4)
- July 2011 (1)
- June 2011 (1)
- May 2011 (5)
- April 2011 (2)
- February 2011 (2)
- January 2011 (2)
- December 2010 (1)
- November 2010 (7)
- October 2010 (3)
- September 2010 (8)
- August 2010 (2)
- July 2010 (4)
- June 2010 (7)
- May 2010 (2)
- April 2010 (1)
- March 2010 (3)
- February 2010 (3)
- January 2010 (2)
- November 2009 (6)
- October 2009 (6)
- August 2009 (3)
- July 2009 (3)
- June 2009 (3)
- May 2009 (2)
- April 2009 (8)
- March 2009 (6)
- February 2009 (4)
- January 2009 (3)
- November 2008 (3)
- October 2008 (7)
- September 2008 (6)
- August 2008 (9)
- July 2008 (9)
- June 2008 (9)
- May 2008 (9)
- April 2008 (8)
- March 2008 (4)
- February 2008 (3)
- January 2008 (3)
- December 2007 (2)
- November 2007 (7)
- October 2007 (1)
- August 2007 (4)
- July 2007 (3)
- June 2007 (8)
- May 2007 (4)
- April 2007 (2)
- March 2007 (2)
- February 2007 (5)
- January 2007 (8)
- December 2006 (1)
- November 2006 (3)
- October 2006 (4)
- September 2006 (3)
- July 2006 (1)
- May 2006 (2)
- April 2006 (1)
- July 2005 (1)
No Comments Yet
Let us know what you think