Presentation is loading. Please wait.

Presentation is loading. Please wait.

Capacity Planning for LAMP Architectures

Similar presentations


Presentation on theme: "Capacity Planning for LAMP Architectures"— Presentation transcript:

1 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Web Builder 2.0 Las Vegas John Allspaw Manager, Operations Flickr.com © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

2 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Capacity planning: “the ability to make snap decisions to spend millions of dollars with not enough information” - Kevin Murphy © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

3 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures It is NOT: Performance tuning Tips and tricks to be “scalable” © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

4 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures It IS: What comes after you’ve made it all “scalable” Making sure that you have enough equipment to handle gradual and bursty traffic © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

5 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Questions to answer © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

6 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures How many of each server “class” should you add as you grow ? Hint: Don’t add too much (too much $$! Ahh!) Hint: Don’t add too little (too much traffic! Ahh!) © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

7 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures How to make it easier to predict the future* ? How to make it easier to justify those predictions ? How to make it easier to predict the future…in the future ?! *You can’t predict the future, but you can try. © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

8 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures The OLD way of doing things was easier A.k.a. “web1.0” Small number of content producers Control over the content Capacity was dictated by the demand for that static/small content Even bbs/communities/ecommerce had relatively predictable growth © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

9 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Today’s way of doing things is harder/fun No control over content (users have control) No control over usage (users have control) © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

10 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Today’s way of doing things is harder/fun Network effect, nonlinear growth (more users/content/contacts/activity mean >> usage) Event-related growth (press, news event, social trend, etc. can affect usage and content) Example: London bombing, tsunami, holidays, etc. © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

11 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Considerations for “social” applications User behavior should guide you with defining capacity metrics (not just server stats) Usage can accelerate, not just grow © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

12 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures How we do it at Flickr © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

13 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Gathering Usage Application-level information (users, photos, activity, etc.) Server-level information (cpu, disk I/O, memory, etc.) We tie the two together © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

14 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures BEFORE we start collecting server stats What resources are peak-driven ? (concurrent use) Ex: photo processed/sec, pages/sec, images/sec, db qps What resources are permanently consumable ? Ex: database space, storage (GB/day) etc. © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

15 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures BEFORE we start collecting server stats What is an average user:consumption ratio ? (example: user: photo) What is the high and low of ratios ? Is the average ratio changing over time ? User:ratios - sometimes users have no photos, only lurk Sometimes users have LOTs of photos, don’t comment/favorite, etc. Ratio rate change: adjust ceiling on user-driven resources © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

16 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Non-linear growth © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

17 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Non-linear growth © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

18 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Linear relationships, though © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

19 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Server and Network statistics Ganglia - (we love ganglia!) Multicast-y goodness SUPER simple to make a graph from any stat Clustering Other custom rrdtool-based stuff MRTG User:ratios - sometimes users have no photos, only lurk Sometimes users have LOTs of photos, don’t comment/favorite, etc. Ratio rate change: adjust ceiling on user-driven resources © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

20 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Photos uploaded/processed/min Avg processing time per minute Avg CPU per minute © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

21 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Gather and record statistics Accept the ‘observer effect’ (it’s worth it) Aggregate your stats across clusters Stacked graphs Totals and averages User:ratios - sometimes users have no photos, only lurk Sometimes users have LOTs of photos, don’t comment/favorite, etc. Ratio rate change: adjust ceiling on user-driven resources © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

22 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Squid client requests (24 hours) (Y axis is req/sec) User:ratios - sometimes users have no photos, only lurk Sometimes users have LOTs of photos, don’t comment/favorite, etc. Ratio rate change: adjust ceiling on user-driven resources © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

23 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Squid LRU reference age Over 24 hours Y axis is “days” So peak has 3.6hours User:ratios - sometimes users have no photos, only lurk Sometimes users have LOTs of photos, don’t comment/favorite, etc. Ratio rate change: adjust ceiling on user-driven resources © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

24 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Find the ceiling of each class/function/server Maximum allowable somethings MySQL : queries/sec before slave lag sets in Apache/php : page requests/sec before total meltdown Squid/memcached : cache churn rate, request rate Storage : disk I/O utilization, storage limit(!) Etc. User:ratios - sometimes users have no photos, only lurk Sometimes users have LOTs of photos, don’t comment/favorite, etc. Ratio rate change: adjust ceiling on user-driven resources © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

25 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Forget benchmarking, use real load Make sure you have a easy mechanism to take servers in and out of production Pull machines from a balanced pool during medium-level traffic (very carefully) Watch and record User:ratios - sometimes users have no photos, only lurk Sometimes users have LOTs of photos, don’t comment/favorite, etc. Ratio rate change: adjust ceiling on user-driven resources © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

26 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Build the infrastructure to make it EASY to measure Obvious things to help this: Load balancing Network segmentation Carve up functions into clusters Don’t let a machine do more than one primary thing (if you can help it) this isn’t for performance! If it makes it faster/better, then bonus! User:ratios - sometimes users have no photos, only lurk Sometimes users have LOTs of photos, don’t comment/favorite, etc. Ratio rate change: adjust ceiling on user-driven resources © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

27 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures For graphs you don’t have raw data for GraphClick - graph digitizer package - $8 US - you pick points on a calibrated image, it spits out tabular data © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

28 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

29 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Once you have: Time history of metrics “Ideal” peak levels (ceiling) Then you can: 3. Predict the future!* © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

30 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Example: Photo Processing © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

31 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Photos uploaded/processed/min Avg processing time per minute Avg CPU per minute © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

32 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

33 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Dirty linear math 25% photos/min 40% photos/min So….take a “ceiling”: 75% photos/min = 6720 photos/hour (but double-check the process time) © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

34 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Conclusion © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

35 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Know your machines and their limits Measure how the site is being used with application-level stats Tie real-world observations to server stats © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

36 Capacity Planning for LAMP Architectures
MGB 2003 Capacity Planning for LAMP Architectures Some Flickr statistics 300M photos, 4 or 5 different sizes Keep ~25M images in cache at any time, ~1M from RAM 2B MySQL queries/day 21k req/sec to memcached 1.2 PT raw disk storage © 2003 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.

37 MGB 2003 © 2003 Microsoft Corporation. All rights reserved.
This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.


Download ppt "Capacity Planning for LAMP Architectures"

Similar presentations


Ads by Google