Upload
salesforce-developers
View
202
Download
3
Embed Size (px)
DESCRIPTION
Can you load 20 million records into Salesforce in under an hour? If not, this webinar is for you. You want to load tons of data into Salesforce. No problem, right? Just use the Bulk API and turn on parallel loading. Think again. Unless you carefully plan the big data loads that you want to break up into parallel operations to achieve maximum throughput, those loads can turn out more like slow, serial loads.
Citation preview
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 intellectual property and other l itigation, risks associated with
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-Q for the most recent fiscal quarter . This 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.
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Time
• One job
• 100 batches
• 10,000 records/batch
• 1M total records
Concurrency Mode Serial
Records Loaded 1 million
Records Failed 0
Run Time 77 minutes
Work Completed 75 minutes
Throughput 13,000 records per minute
Degree of Parallelism 0.97
Key Problem Degree of parallelism explicitly limited to ~1.
Solution Explore parallel load for increased throughput.
0
50000
100000
150000
200000
250000
300000
350000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Serial
Serial Run
• Low degree of parallelism
Degree of Parallelism
Thro
ughput R
ecord
s/M
in
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Thread
Time
• One job
• 100 batches
• 10,000 records/batch
• 1M total records
Concurrency Mode Parallel
Records Loaded 396,600
Records Failed 603,400
Run Time 17 minutes
Work Completed 3 hours 15 minutes
Throughput 22,000 records per minute
Degree of Parallelism 11.5
Key Problem Lock Exceptions. Server worked significantly harder but no increase in throughput.
Solution Run the load in serial mode or manage locks.
0
50000
100000
150000
200000
250000
300000
350000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Serial
Parallel Run 1
• High degree of parallelism
• Low throughput due to locks
Degree of Parallelism
Thro
ughput R
ecord
s/M
in
Parallel 1
Concurrency Mode Parallel
Records Loaded 1 million
Records Failed 0
Run Time 3 minutes and 30 seconds
Work Completed 1 hour
Throughput 320,000 records per minute
Degree of Parallelism 19
Key Problem None
Solution n/a
0
50000
100000
150000
200000
250000
300000
350000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Serial
Parallel Run 2
• High degree of parallelism
• High throughput
Degree of Parallelism
Thro
ughput R
ecord
s/M
in
Parallel 2
Parallel 1
Concurrency Mode Parallel
Records Loaded 1 million
Records Failed 0
Run Time 4 minutes
Work Completed 1 hour
Throughput 250,000 records per minute
Degree of Parallelism 16.5
Key Problem Minimal overhead due to locks
Solution Remove all unnecessary locks
0
50000
100000
150000
200000
250000
300000
350000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Serial
Parallel Run 3
• High degree of parallelism
• High throughput
Degree of Parallelism
Thro
ughput R
ecord
s/M
in
Parallel 2
Parallel 3
Parallel 1
0
50000
100000
150000
200000
250000
300000
350000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Serial
Controlled Feed Run
• Reduced parallelism
• Expected throughput
Degree of Parallelism
Thro
ughput R
ecord
s/M
in
Parallel 2
Parallel 3
Controlled Feed
Parallel 1