Is Real Time more important than Right Time? When is it the right time for you to be real time?

1. Copyright © 2010, SAS Institute Inc. All rights reserved.
When is the right time for real-time?

2. 2
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Profile
An experienced analytics specialist with extensive
experience in risk analytics, marketing analytics and
social media analytics. Extensive experience in
developing new techniques and models to achieve
business objectives. Former SAS Global Forum Section
Chair.

3. 3
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Real time…?
• Do I need real time capabilities?
• What is real time for me?
• Right time vs real time
• Free lunch? No way
• Open Source = Free Lunch?
• Real Time, Right Time, Robust Returns

4. 4
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Do I need real time capabilities?
The need for real time capabilities is driven by the
business.
Model
Environment
Impact

5. 5
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Model
Type of Model Business Examples
Price Models Buying Club, One-stop, low-price shopping Fee for advertising, Razor and blade
Convenience Models One-stop, convenient shopping, Comprehensive offering, Instant gratification
Commodity-Plus
Models
Low-price reliable commodity, Mass customised commodity, Service-wrapped commodity
Experience Models Experience selling, Cool brands
Channel Models Channel maximisation, Quality selling, Value-added reseller
Intermediary Models Market aggregation, Open market-making, Multi-party market aggregation
Trust Models Trusted operations, Trusted product leadership, Trusted service leadership
Innovation Models Incomparable products, Incomp
Linder and Cantrell’s (2000) categorization of business models

6. 6
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Environment

7. 7
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Impact (Courtesy of Alfrid from The Hobbit)
Where a little gold coin has better effect than a Wizard’s
wand….
https://www.youtube.com/watch?v=-fcJm1Slk2E

8. 8
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
What is real time for me?
In real time analytics, the definition of real time is the time
interval between the reception of data, performing the
analysis and finally taking the actions that will be relevant.
Milliseconds
Seconds
Minutes
Hours
Algorithmic Trading
Fraud Detection
E-Commerce
Location Based
Marketing
Online Retailers

9. 9
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Right Time Vs Real Time
Even with real time, the more important thing is to have
the right timing.
Right
Time
Real
Time

10. 10
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Real Time Marketing - Mall
Pre-Store In-Store Post-Store
When should I target?
What should I target?
Target the approaching
customer with an offer
that is in a shop 10
meters away.
Target an in-store
customer with an offer
that is store based.
Offer a return purchase
coupon.

11. 11
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Free Lunch? No Way

12. 12
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Free Lunch? No Way
mea
upd
quic
Adv
Proc
easi
prov
tion
righ
spe
man
norm
mor
can
cust
to h
Uns
prov
extr
modelsin a single,
centrally managed
intuitive interface.

13. 13
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Open Source = Free Lunch?
Hardware Provisioning
A common question received by Spark developers is how to configure hardware for it. While the right hardware will depend on the situation, we make the following
recommendations.
Storage Systems
Because most Spark jobs will likely have to read input data from an external storage system (e.g. the Hadoop File System, or HBase), it is important to place it as close
to this system as possible. We recommend the following:
If at all possible, run Spark on the same nodes as HDFS. The simplest way is to set up a Spark standalone mode cluster on the same nodes, and configure Spark and
Hadoop’s memory and CPU usage to avoid interference (for Hadoop, the relevant options are mapred.child.java.opts for the per-task memory and
mapred.tasktracker.map.tasks.maximum and mapred.tasktracker.reduce.tasks.maximum for number of tasks). Alternatively, you can run Hadoop and Spark on a
common cluster manager like Mesos or Hadoop YARN.
If this is not possible, run Spark on different nodes in the same local-area network as HDFS.
For low-latency data stores like HBase, it may be preferrable to run computing jobs on different nodes than the storage system to avoid interference.
Local Disks
While Spark can perform a lot of its computation in memory, it still uses local disks to store data that doesn’t fit in RAM, as well as to preserve intermediate output
between stages. We recommend having 4-8 disks per node, configured without RAID (just as separate mount points). In Linux, mount the disks with the noatime option
to reduce unnecessary writes. In Spark, configure the spark.local.dir variable to be a comma-separated list of the local disks. If you are running HDFS, it’s fine to use the
same disks as HDFS.
Memory
In general, Spark can run well with anywhere from 8 GB to hundreds of gigabytes of memory per machine. In all cases, we recommend allocating only at most 75% of
the memory for Spark; leave the rest for the operating system and buffer cache.
How much memory you will need will depend on your application. To determine how much your application uses for a certain dataset size, load part of your dataset in a
Spark RDD and use the Storage tab of Spark’s monitoring UI (http://<driver-node>:4040) to see its size in memory. Note that memory usage is greatly affected by
storage level and serialization format – see the tuning guide for tips on how to reduce it.
Finally, note that the Java VM does not always behave well with more than 200 GB of RAM. If you purchase machines with more RAM than this, you can run multiple
worker JVMs per node. In Spark’s standalone mode, you can set the number of workers per node with the SPARK_WORKER_INSTANCES variable in conf/spark-
env.sh, and the number of cores per worker with SPARK_WORKER_CORES.
Network
In our experience, when the data is in memory, a lot of Spark applications are network-bound. Using a 10 Gigabit or higher network is the best way to make these
applications faster. This is especially true for “distributed reduce” applications such as group-bys, reduce-bys, and SQL joins. In any given application, you can see how
much data Spark shuffles across the network from the application’s monitoring UI (http://<driver-node>:4040).
CPU Cores
Spark scales well to tens of CPU cores per machine because it performes minimal sharing between threads. You should likely provision at least 8-16 cores per machine.
Depending on the CPU cost of your workload, you may also need more: once data is in memory, most applications are either CPU- or network-bound.
The hardware for
50TB storage
is SGD$XXXXXXX.
Plus, you need
a team of
5 Hadoop
Specialists.
What about
the deployment
period?

14. 14
Copyr ight © 2012, SAS Institute Inc. All rights reser ved.
Real Time, Right Time, Robust Returns
• Several Retail Space Companies in Singapore have
started reaping the benefits of doing real time analytics.
• Telcos have been progressing in this area through
Geo-Fencing Technology to market relevant offers.
• Banks also started offering some form of real time
marketing.
• Most companies have reported gains in the deployment
of real time technology.

15. Copyright © 2010, SAS Institute Inc. All rights reserved.
Thank you!