SynapSense

1
SynapSense www.SynapSense.com
2
WSN Market Evolution

• WSN 2nd Generation
• The Commercial Era
• Customer Requirements
• Hardware platform flexibility
• Accelerated application development time
• Enable broadly varying sensor networks
• Rapid adoption of continuously emerging
  technology improvements
• Result Complete, commercial solutions

WSN 1st Generation The Research
Era Issues Incomplete Solutions Subject to the
Tech Treadmill Many non-standard
platforms Difficult to build, deploy and
manage Domain-specific languages Result Few
complete solutions, fragmented value-chain
2000 2007 2010
3
System Overview
EAI Web Services, SNMP, BACnet ModBus
!
5
Data and Control Bus
Gateway
Storage
4
3
Data Integration
Data Analysis Software
Data Analysis
XML, SCADA
4
The Ecosystem Challenge
Sensors
Visualization
Database
WSN Packages
Installation
Service Support
WSN boards
Data Collection
Business Intelligence
Bag of Parts
Very Difficult
Expensive Intrusive
5
SynapSense Solution OverviewArchitectural
Overview
6
Sensor Types
Chilled Water BTU
Branch Circuit Monitoring Power
Metering
Leak Detection
Air Pressure
Rack Thermals
Room RH Dew Point
Door Status
7
SynapSense Solution Overview
CRAH Supply Return Air Temp Supply RH Chilled
Water Supply Return Temp Fan Power
PDU Current, Voltage Power
Rack Temperature RH Current Power
Sub-floor Pressure Differential
Chilled Water Flow (Captured at CRAH)
8
LiveImaging SynapSoft 4.0
2D Thermal Map
2D Pressure Map
2D Humidity Map

• Ground-breaking 2D maps use your real-time data
  center analytics to create a visual description
  of your environment.
• Live Environmental Imaging is a special service
  for SynapSense customers.

9
Towards a Greener Data Center Tools Techniques

UCCSC 2008, UC Santa BarbaraGregory BellActing
Chief Technology Architect, IT DivisionLawrence
Berkeley National LaboratoryJuly 21, 2008
10
Motivation

• Energy efficiency core DOE mission
• several LBNL researchers study power consumption
  of IT equipment (data centers, servers, power
  supplies, cooling, UPS technology)
• More pragmatically data center resources are
  increasingly scarce!
• power, space, cooling all in short supply
• Why? Significant growth in cluster computing
• 40 of LBNL scientific teams need or use clusters
• 70 interested in cycles on shared cluster

10
11
Motivation scientific computing

• Demand for computing cycles is
• growing rapidly on many campuses, not just LBNL
• the direct expression of a broad scientific need
• ITs primary data center runs out of capacity
  between 2010 and 2012
• depending on cluster growth rate (7 - 20)
• Strategically, we must maximize efficiency of
  current data center
• even as we locate new space

11
12
The story of a data center 50B-1275

• Several decades old
• 5000 sq ft
• 18 raised floors
• .5MW total power consumption
• Combination of clusters and business systems
• clusters account for almost all new demand
• Unusual cooling system
• 7 down-shot CRACs for under-floor supply
• supplementary overhead supply (chiller coil
  fans)

13
The story of a data center 50B-1275

• Starting summer of 2007, IT began a data center
  efficiency effort
• In collaboration with researchers in EETD, we
  perform a series of engineering studies
• documenting electrical, mechanical systems
• measuring loads capacities performing CFD
• calculating efficiency
• identifying opportunities for improvement
• deploying wireless sensor system (more later)

14
What we learned

• The data center was overcooled
• but this is the norm
• Abundant low-hanging fruit
• simultaneous humidification dehumidification
• blanking panels missing
• excessive number of perforated floor tiles
• air-flow pathologies (mixing, short-circuiting)
• The data center was relatively efficient!
• none of the clusters backed by UPS

15
Taking steps to increase efficiency

• Improved airflow
• overhead plenum converted to hot-air return
• supplementary cooling directed under floor
• floor-tile tuning
• blanking panels, curtains, etc
• Virtualization
• Water cooling
• Wireless environmental monitoring system

16
Steps to increase efficiency

• Improved airflow
• overhead plenum converted to hot-air return
• supplementary cooling directed under floor
• floor-tile tuning
• blanking panels, curtains, etc
• Virtualization
• Water cooling
• Wireless environmental monitoring system
• focus of remainder of talk

17
The importance of visualization

• systems network administrators have tools for
  visualization
• useful for debugging, benchmarking, capacity
  planning, forensics
• data center managers have comparatively poor
  visualization tools

17
18
One form of visualization CFD
Images ANCIS
19
Visualization and real-time monitoring

• We installed a wireless sensor net from
  SynapSense
• 200 monitoring points
• temperature, humidity, under-floor pressure,
  current
• For the first time, we have a detailed
  understanding of environmental conditions in the
  data center
• real-time and historical graphs
• underlying database
• SynapSense is a technology transfer success story
  from UC Davis
• CTO and co-founder is Raju Pandey, also Computer
  Science professor at UCD
• http//www.synapsense.com
• You cant control or manage what you dont
  measure.

20
SynapSense hardware

• wireless sensor network
• self-organizing nodes
• 802.15.4 (not 802.11)
• multi-hop routing
• non-invasive installation
• 2 internal 6 external sensors per node
• we measure temp., humidity, pressure, current
• we could measure liquid flow, liquid presence,
  particle count

Image SynapSense
21
Learning from the sensors

• lesson 1 eat your spinach!
• effect of adding one 12 blanking panel to the
  middle of a rack
• conventional wisdom is correct blanking panes
  are essential
• other data center clichés also borne out
• eliminate leaks in floor
• manage floor tile permeability

Top of rack
Middle of rack
22
Visualizing impact of site modifications

• Graphing impact of major maintenance (redirection
  of overhead cold air supply underfloor) on air
  pressure
• in some areas, under-floor pressure increased by
  almost 50
• impact varies according to distance from new air
  supply

23
Visualizing impact of maintenance
Under-Floor Pressure

• each CRAC turned off in turn, for service
• hot spots monitored during maintenance
• two-fold benefit
• real-time feedback
• enhanced knowledge of data center redundancy
  characteristics
• provisionally turn off one or more CRAC units?

Rack-Top Temperatures
24
Feedback on floor-tile tuning
Before

• in the course of one day
• 24 perforated floor tiles removed
• 6 floor tiles converted from high- to low-flow
• 4 floor tiles converted from low- to high-flow

After
25
Feedback on floor-tile tuning

• with instrumentation, we can observe results in
  real time
• when airflow is restricted,
• under-floor pressure increases
• rack-top temperatures decrease
• without monitoring and visualization, this
  process is guesswork
• how many tiles to remove?

Under-Floor Pressure
Rack-Top Temperatures
26
Focus on a single sensor

• iterative process too cold, too hot, just right
• monitor with laptop in data center for convenience

27
LiveImaging heat-map movies
27
28
Data Center Value
29
SynapSense 3-Step Process to Adaptive Cooling
Air Flow Management
2
Visibility
Adaptive Cooling Control
1
3
30
Environmental Baseline ResultsSynapSense
Thermal LiveImaging
Baseline
C
H
H
C
No Containment
High Degree of Air Mixing
High Inlet Temperatures
31
SynapSense Solution OverviewCold Aisle
Containment
32
Environmental Baseline ResultsSynapSense
Thermal LiveImaging
After Containment
C
H
H
C
Cold Aisles Contained
No Change to CRAHs
Overcooled
33
Environmental Baseline ResultsSynapSense
Thermal LiveImaging
With Control
C
H
H
C
Controlled CRAHs
ASHRAE Inlet Temperatures
Increased Return Temperatures
34
Environmental Baseline ResultsSynapSense
Thermal LiveImaging
Baseline
After Containment
With Control
C
H
H
C
C
C
H
H
C
H
H
C
35
One Solution Three Vectors
Balance Cooling Capacity
Optimize
Air Pressure Flow CRAC/CRAH
Utilization
Cooling
Optimization
Improve Energy Management
UPS Capacity Planning Energy
Baseline Metrics PUE
/ DCIE
Power
Balance
Synergize IT Facilities One
Dashboard Integrated Tool Suite
Optimized Capacity Planning Balanced IT
Infrastructure
Operations
Efficiency
36
Qualified for Utility Rebates
PGE Once improvements made, customer receives
incentive of .08 per kWh saved, up to half cost
of improvements
SVP will rebate 0.18/kWh saved up to 80 of the
solution cost with proven savings

• SMUD will provide incentive 0.14/kWh saved not
  to exceed 30 of project cost or 50,000