The document discusses how to scale applications for enterprise use on the Force.com platform. It recommends estimating user and data volume growth, designing for selectivity and data reduction, using the right data architecture like indexes and skinny tables, and performance testing applications at scale before customers encounter issues. The document also previews upcoming Force.com capabilities for improved data management and archiving that can help with scalability.

1. Avoid Growing Pains: Scale your App
for the Enterprise
Randy Case
ISV Technical Evangelist
@randyscase

2. Safe Harbor
Safe harbor statement under the Private Securities Litigation Reform Act of 1995:
This presentation may contain forward-looking statements that involve risks, uncertainties, and assumptions. If any such uncertainties materialize or if any of
the assumptions proves incorrect, the results of salesforce.com, inc. could differ materially from the results expressed or implied by the forward-looking
statements we make. All statements other than statements of historical fact could be deemed forward-looking, including any projections of product or service
availability, subscriber growth, earnings, revenues, or other financial items and any statements regarding strategies or plans of management for future
operations, statements of belief, any statements concerning new, planned, or upgraded services or technology developments and customer contracts or use
of our services.
The risks and uncertainties referred to above include – but are not limited to – risks associated with developing and delivering new functionality for our service,
new products and services, our new business model, our past operating losses, possible fluctuations in our operating results and rate of growth, interruptions
or delays in our Web hosting, breach of our security measures, the outcome of any litigation, risks associated with completed and any possible mergers and
acquisitions, the immature market in which we operate, our relatively limited operating history, our ability to expand, retain, and motivate our employees and
manage our growth, new releases of our service and successful customer deployment, our limited history reselling non-salesforce.com products, and
utilization and selling to larger enterprise customers. Further information on potential factors that could affect the financial results of salesforce.com, inc. is
included in our annual report on Form 10-K for the most recent fiscal year and in our quarterly report on Form 10-Q for the most recent fiscal quarter. These
documents and others containing important disclosures are available on the SEC Filings section of the Investor Information section of our Web site.
Any unreleased services or features referenced in this or other presentations, press releases or public statements are not currently available and may not be
delivered on time or at all. Customers who purchase our services should make the purchase decisions based upon features that are currently available.
Salesforce.com, inc. assumes no obligation and does not intend to update these forward-looking statements.

3. Agenda
– Introduction
– Thinking at enterprise scale
– Force.com database architecture and tuning
– Architecting your application for scale
– Testing your application at scale
– Peek into the future
– Question and Answers

4. Understand application and user requirements
Estimate volumes
at enterprise scale
Operational vs.
Historical Data
Understand where
scale will be used

5. Estimate the growth and volume of data
Can you achieve a
steady state on
volume?
Where does growth
come from?
How fast will data
volume increase?

6. Design your app for scale
Understand the Test, Test, Test, …
impact areas of LDV
Always think big
(be reasonable)!

7. Force.com Database Architecture and
Tuning

8. Force.com Database Architecture
Shared Database
Data and Pivot Tables

9. Go
Check user
Visibility
Check filter
selectivity
Write query-based
on results
of pre-queries
Execute query
User
Visibility
# of rows that
the user can
access
=
Filter
Selectivity
How specific
= is this filter?
Stop
Multi-tenant Query Optimizer
Multi-tenant
Optimizer
Statistics

10. Architecting Your Application for Scale
What you need to do

11. The Key to Performance
Selectivity
Data Reduction
§ Reduce the amount of data accessed during query.
§ Make query parameters specific as possible.
§ Less data means better performance
§ Consider archiving in initial design
§ Many features to facilitate archiving process.

12. Reports
1. Use focused report filters 2. Pre-aggregate the data
3. De-normalize the data 4. Use a data warehouse

13. API Best Practices
1. No Apex triggers and workflow 2. Operation choice
3. Record sizes 4. Optimize sharing
if possible

14. API Best Practices
1. Delta-only loads 2. Parent record locking
3. Keep large volumes
out of DB
4. Database statistics

15. Sharing
• Used by all features and can have huge impact with LDV.
• Data accessibility is governed by:
– Organization-wide defaults
– Role hierarchy
– Sharing rules
– Criteria-based sharing
– Manual shares
– Programmatic sharing
– Implicit sharing

16. 1. Avoid over-protection of data
3. Avoid skew 4. Don’t nest groups
too deeply
Sharing Best Practices
2. Leverage the right
tools for the Task

17. SOQL Best Practices
§ Make queries selective
§ Use alternative approach
§ Utilize Apex to do the work
§ Decompose relationship queries
§ Some use cases that will cause issues:
§ Performing large data loads and immediate query
§ LIKE with leading wildcards “%Acme”
§ Use of NOT and “!=“
§ Complex queries with multiple joins
§ Data in the Recycle Bin
Query Resource Feedback
Parameter Pilot

18. DB
SEARCH
INDEX
DB processing
includes sharing
access check
Millions of
records
Query across
entire index
for matches to
search string
Record IDs of
matches to
query are sent to
DB for
processing

19. Search Best Practices
1. Keep search text specific 2. Avoid complex sharing
3. Single object searches
4. LDV data changes

20. Architecting Your Application for Scale
How Salesforce may help you

21. What are Indexes?
ID First Name Last Name
1000001 John Smith
1000002 Jane Jones
1000003 Buzz Baker
ID First Name Last Name
1000003 Buzz Baker
1000002 Jane Jones
1000001 John Smith

22. Index Rules
• Must be below thresholds to be useful:
– Custom Indexes: Lower of (10% of 1M + 5% over 1M) or 333,333.
– Standard Indexes: Lower of (30% of 1M + 15% over 1M) or 1M.
• Modified by “AND” and “OR”
– AND – Below 20% of total records or 666K for custom index
– OR – Below 10% of total records or 333K in summation of all indexes.
• Predicates that won’t use indexes
– Not equal to (i.e. “!=“)
– NOT(i.e. NOT IN)

23. Skinny Tables
• Reduced I/O
• Better cache Utilization
• Reduced Joins
• Invisible to End users
• Supported in reports, list views and
SOQL queries
• All fields must be in skinny table

24. Testing Your Application at Scale

25. What Is Performance Testing?
Response Time Based Testing
Confirm that a set of process flows perform in
such a manner as to give the end user a
positive experience
User Load Based Testing
Confirm that a set of process flows will be able
to support a particular user load.

26. Candidates for Performance Testing
• Areas effected by large data volumes
– Custom code (VF, Apex)
– Reports, List views
– Page detail with possible skew
• Usage Patterns
– Large number of records
– Large number of users
– Large number of transactions

27. What is response time based testing?
• This is functional and execution performance testing at scale.
– Functional testing
• Ensure all app functionality executes without error
• Ensure UX is optimal at large data volumes
– Execution performance testing
• Understand relative response times
• Focus on improving relative performance

28. What is user load based testing?
Simulate Users
Load Generators
Load Controller
Load Metrics

29. Peek into the Future

30. Future Enhancements
Big Objects Fileforce
API
• Efficient storage for non-transactional data • Unstructured and Semi-structured Blob Store
Data
Pipelining
Data
Archiving
• Highly-scalable batch processing Programmatic data lifecycle from live to archive state

31. Dreamforce Sessions
• Introducing Force.com Data Management Services
– Wednesday, 4:00 PM - 4:40 PM
– Century Theaters
– Theater 5
• Get Out of Cold Storage, Get into Data Archiving on Salesforce
– Wednesday, 11:00 AM - 11:40 AM
– Century Theaters
– Theater 2

32. Summary
• Develop your app with scale in mind.
• Use best practices when building.
– Keep data volumes to a minimum.
– Optimize and tune programmatic components
• Test for scale before your customer does.

33. Resources
• Developer Website
• The ISV team will help you!

34. + “Forgot the ring” =
+ “Application
Scalability” =
+ “What work?” =

35. Check out the new Partner Community
https://partners.salesforce.com/

36. Connect with Partners in the Partner Zone
The Westin Hotel, Market Street
2nd Floor – Metropolitan Ballroom
INNOVATE with the leading technology
• Demos of new Salesforce technology
CONNECT with members of the partner community
• Partner Community Theater
• Networking areas
• Welcome reception and daily lunch service
GROW your business with resources
• 70+ partner-specific sessions
• ‘Ask the Experts’ consultation stations

37. AppBash 2014 on Wednesday Night!