Date post: | 10-Apr-2018 |
Category: |
Documents |
Upload: | rafael-bonutti |
View: | 248 times |
Download: | 0 times |
of 29
8/8/2019 Yi Cloud10 Pres
1/29
Reducing Costs of Spot Instances via
Checkpointing in the Amazon Elastic
Compute Cloud
Sangho Yi, Derrick Kondo, and Artur Andrzejak
- at CLOUD 2010 -
Sangho Yi
INRIA Grenoble, France([email protected])
8/8/2019 Yi Cloud10 Pres
2/29
Outline
Motivation and background
Spot Instances on Amazon EC2
Possible checkpointing strategies Performance evaluation
Conclusions and future work
8/8/2019 Yi Cloud10 Pres
3/29
Motivation:
Market-based Pricing in Cloud Amazon released Spot instances in Dec.
2009.
Current price depends on users bids, andavailable computing resources.
(based on the supply and demand relationship)
30~50% price of normal instances.
In-bid users get instances.
Out-of-bid users lose their instances.
8/8/2019 Yi Cloud10 Pres
4/29
Pricing in Amazon EC2
(normal instances on demand)Virtual MachineInstance Type
Linux
Cost/hour (USD)
MS Windows
Cost/hour (USD)
Standard Small 0.095 0.12
Standard Large 0.38 0.48
Standard Extra Large 0.76 0.96High CPU - Medium 0.19 0.29
High CPU - Extra Large 0.76 1.16Data Transfer
Type
Cost/GB-Month
(USD)
Inbound Transfer 0.10
First 10 TB 0.17
Next 10-50 TB 0.13
Next 50-150 TB 0.11
Over 150 TB 0.10
** Pricing on eu-west (Ireland), 2010
8/8/2019 Yi Cloud10 Pres
5/29
Pricing in Amazon EC2
(spot-instances bidding system)
Source: http:/cloudexchange.org
8/8/2019 Yi Cloud10 Pres
6/29
An Example of Failures as a
Function of Users Bid
Users bid = 0.081
Availability Availability Availability
Failure Failure
Current Price
8/8/2019 Yi Cloud10 Pres
7/29
PDF of Failure arrivals
8/8/2019 Yi Cloud10 Pres
8/29
Goal
Understand impact of checkpointing methods onspot instances in terms ofmonetary cost and
execution time
0
Price
Time (hour)1 2 3 4 5 6 7 8 9 10
0.1
0.20.3
0.4Bid
availability (5) availability (3)failure (2)
8/8/2019 Yi Cloud10 Pres
9/29
Possible Checkpointing
Strategies In terms of taking points
The optimal case
Without checkpointing
Hourly checkpointing
Rising edge-driven checkpointing
Checkpointing with adaptive decision whetherto skip or take a checkpoint
Combinations of aboves
8/8/2019 Yi Cloud10 Pres
10/29
Characteristics of Strategies
Hourly
Pay as you go (with checkpointing per hour)
Rising edge-driven Takes when price goes up (except for out-
of-bid cases)
Adaptive taking point decision
Has overhead for decision, also thedecision may not work well in some cases.
8/8/2019 Yi Cloud10 Pres
11/29
Hourly vs. Edge-driven
tc
tc
tc
Time (minute)0 60 120 180
Failure
(without
payment)
Recovery
Pay per hour Pay per hour Pay per hour
Task execution Task execution Task execution
Task
execution
Time
Price
fora
s
potinstance
Users bid
Available duration Available duration
Failure
Recovery
: checkpoint
: rising edge
8/8/2019 Yi Cloud10 Pres
12/29
Adaptive Taking Point Decision
Based on time series analysis,
it compares the expected costs at a certain point
when Taking a checkpoint
Skipping a checkpoint
Then, selects the lower-cost option.
8/8/2019 Yi Cloud10 Pres
13/29
System Model
8/8/2019 Yi Cloud10 Pres
14/29
To take or skip, that is the question
tc
tc
Time0
Failure
Recovery
(short rollback)
Task execution Task execution Task execution
tc
Time0
Failure
Recovery
(long rollback)
Task execution Task execution
(a) When taking a checkpoint at the current time
Current time
Current time
(b) When skipping to take a checkpoint at the current time
8/8/2019 Yi Cloud10 Pres
15/29
Cost analysis on the choice
8/8/2019 Yi Cloud10 Pres
16/29
Delayed termination a way of
exploiting Amazons policy
Time (minute)0 60 120 180
Failure(out-of-bid)
Pay per hour Pay per hour Pay per hour
Task execution Task execution Task execution
Task
execution Without payment
Time (minute)0 60 120 180
End of execution
(when user stops)
Pay per hour Pay per hour Pay per hour
Task execution Task execution Task execution
Pay last partial-hour
(a) When a users spot instance is out-of-bid
(b) When a user stops using spot instance
Time (minute)0 60 120 180
Users bid
Available durationPrice for spot instance
8/8/2019 Yi Cloud10 Pres
17/29
Performance evaluation
Trace-based simulation
http://spotckpt.sourceforge.net
42 spot instances (now Amazon has 64)
24 (12x2) different checkpointing strategies
8/8/2019 Yi Cloud10 Pres
18/29
Checkpointing strategies
OPT: The optimal case
NONE: Without checkpointing
H: Hourly
E: Rising edge-driven
AH: H with adaptive decision
AE: E with adaptive decision
H+E, H+AE, AH+E, AH+AE
AF(10): adaptive decision every 10 mins
AF(30): adaptive decision every 30 mins
8/8/2019 Yi Cloud10 Pres
19/29
Results #1: Monetary cost
Total cost according to Users bid on two spotinstances
8/8/2019 Yi Cloud10 Pres
20/29
Results #2: Execution time
Total execution time according to Users bid onthe spot instances.
8/8/2019 Yi Cloud10 Pres
21/29
Results #3: Normalized one
Results #4: Benefit of using DT
8/8/2019 Yi Cloud10 Pres
22/29
Results #5: Best ones on the
mean price bidding
Hourly checkpointing shows good performance.
Fine-grained checkpointing shows good performance
sometimes.
Adaptive decision does not show significant impact onthe results.
8/8/2019 Yi Cloud10 Pres
23/29
Summary
Checkpointing strategies have ~50% higheroverhead (execution time * monetary cost)
than the optimal case.
Thus, better checkpointing strategies arerequired to reduce both cost and time.
8/8/2019 Yi Cloud10 Pres
24/29
Current and Future work
1) Finding relationship between past andfuture price of spot instances
2) Developing efficient checkpointingmethod to minimize monetary cost and
execution time (to appear @ EKC 2010)
3) Providing decision model to determineusers bid, given reliability and performance
constraints (to appear @ MASCOTS 2010)
8/8/2019 Yi Cloud10 Pres
25/29
Thanks! / Questions?
Lets walk (or work) on the Cloud!
8/8/2019 Yi Cloud10 Pres
26/29
** On-going work on 2)
Sangho Yi and Derrick Kondo How CheckpointingCan Reduce Cost of Using Clouds?, EKC 2010, July,
2010. (accepted)
We used the known characteristics of AmazonsSpot Instances, and better checkpoint decision
mechanism.
The proposed scheme shows much less
overhead for most cases. (~20% overhead)
8/8/2019 Yi Cloud10 Pres
27/29
8/8/2019 Yi Cloud10 Pres
28/29
** On-going work on 2)
8/8/2019 Yi Cloud10 Pres
29/29
Lets Make Our Cloud Better!
On a Cloud Vendors point of view,
TODO
(efficiently) Adapt management strategies to fullyutilize the limited (very-large) resources (with less
cost)
Better
Utilization,Less Cost
More
Cloud Users
MoreMoney
More Investment
on R&D