SlideShare a Scribd company logo
ACHIEVING	HIGH	LOAD	IN	AD	TECH
Peter	Milne,	Technology	Architect								 @helipilot50							 helipilot50
AGENDA
About	Adform
High	load	definitions
Demand	Side	Provider	(DSP)
Data	Management	Platform	(DMP)
DevOps
Data	Storage
Uncle	Pete’s	advice
THIS	IS
ADFORM
The	leading	open	and	independent	full	stack	Ad	technology	platform	
Innovating	the	automation	of	buying	and	selling	of	advertising
800	industry	professionals	
22	offices	in	19	countries
330	Development	– 15	Data	Scientist	– 55	Ops,	DBAs,	Infrastructure,	etc
Founded	in	Denmark	in	2002
Under	the	leadership	of	3	code-writing	entrepreneurs
HIGH	LOAD
Shinagawa	Railway	Station	– Tokyo,	Japan
12	December	2014	08:22	AM
Velocity,	Scale,	Availability
VELOCITY	is LATENCY
Latency	is	a	time	interval	between	the	
stimulation and	response.
Latency
Time
Start EndI/O
SCALE	is THROUGHPUT
Throughput	is	the	rate	of	production
Similar	to	Power:	
The power	of	a	system	is	
proportional	to	its	throughput
P	=
W
∆𝑡
Knowledge	=
Work
Money
Money	=
Work
Knowledge
AVAILABILITY	is ALWAYS	ON
Always	on
Telephone
Electrical	power
Water	supply
Sewer
…or	for	the	young
REAL	TIME	BIDDING	(RTB)
WHAT	IS	RTB
1. Publisher	sells	space	
called	an	Impression
2. Exchange	conducts	
and	Auction
3. Advertisers	bid to	fill	
this	space
RTB	- AUCTION
Sell	Side	(supply) Buy	Side	(demand)
OpenRTB Server
Auction	Service
Sell	Side	Platform	
(SSP)
Demand	Side
Platform
(DSP)
OpenRTB Client
Bidder	Service
Advertiser	
Ad	Server
Agency,	Trade	Desk	
or	In-house	team
Publisher	
Ad	Server
Publisher
Ad	Network	
(working	for	a	
publisher)
Real	Time	Bidding	(RTB)	is	the	kernel	of	programmatic	advertising
DSP	SIZE,	SLA	&	VALUE
Size:	8 billion	cookies
Availability:	24/7/365
Bid	requests:	1.2	million	per	sec
Bid	latency:	~50ms
Bid	percentage:	32%
Bid	wins:	~2%
Value:	
Auction	total:	~1	million	€	per	sec
Bid	Wins:	~8000€ per	sec
DSP	TECHNOLOGY	– BIDDER
Load	balancer:	
12x	Nginx
Bidders:	
102	in	.Net
Segment
Price
Bid
Budget	spend
Research	data
Pricing	APIAerospike	clusterBudget	ServerThrottler
8	x	5.5TB
OPTIMIZATION	RESULTSBUDGETS
COOKIE	PROFILES
Reads/s
Metadata	Servers
Bidder	ServersLoad	Balancer	
(NGINX)
SSP1
SSP2
SSP3
800-500k/s 8k/s
per	server
230k/s
Bids/s
DSP	TECHNOLOGY	- COOKIE	PROFILE	STORE
Secret	weapon:	Aerospike
800-900k/s reads/s,	100k/s	writes
<	1ms	latency	to	read	profile
8	billion	cookies	in	8	servers	
HA with	redundancy
Moved	from	Cassandra	3 years	
ago	– Why?
Easy to	use
Value	for	money	
150	servers	->	8
RTB	PROBLEMS	FACED
Problem:	High	latency	to	find	the	correct	Ad	(line	
item)	to	bid	with.
Cause:	.Net Dictionary	implementation	pushed	
beyond	it’s	design
Implication:	Missed	the	opportunity	to	bid	à
loss	of	revenue
Solution:	Go	back	to	Computer	Science…
Replace	Dictionary	with	fixed	byte	Array	and	
binary	search
Result:	reduced	latency	producing	2x capability
DATA	MANAGEMENT	PLATFORM	
(DMP)
WHAT	IS	A	DMP
The	ways	how	we	attribute	profiles	are	following:
Pixels	on	web	pages
Cookies
Server	to	server
Custom	rules	specified	by	data	providers
Cross-device	algorithms
Look-a-like	algorithms
A	solution	to	store	and	attribute	anonymous	user	profiles	- 1st party	and	3rd party	
Profiles	are	analyzed	and	grouped	into	Audiences	and	Segments	(sets).
Segments,	and	their	cookies,	used	for	Real	Time	Bidding.
DMP	SIZE
4b	cookies	– 50%	map	to	external	
providers
(Google,	AppNexus,	MediaMath,	etc)
60% of	cookies	have	1st party	data	
(browser,	language,	country	etc)
130b	cookie-segment	pairs
relations	between	profiles	by	their	attributes
70k	of possible	cookie	attributes
about	1k	form	data	providers
Cookie profile	updates	50k/s
100k/s	spikes
2x	growth	every	year
DMP	ARCHITECTURE
Restful	API	(API	First	approach)
Micro-services
OAuth	2
Lambda-like	architecture	for	
data	flow
Kappa-like	architecture	for	all	
real-time	data	flows
DMP	TECHNOLOGY
Polyglot	environment
.NET – for	API	and	some	foreground	processing
Java/Scala – everywhere	where	we	need	scale
Java	Script/React– UI
Spark – Parallel	Processing
Polyglot	persistence
Aerospike – for	fast	read	and	write	of	user	profiles
MongoDB – for	metadata	and	fast	UI
Vertica – Analytics	database
MSSQL – for	places	where	we	need	price	accuracy
Ceph (distributed	storage)	– for	files
HDFS	– “Raw”	data	files
Parquet – Efficient	file	format
Redis – for	statistics	and	probabilistic	counters
Kafka/Confluent – for	data	streams
Wide	number	of	technologies:	“right	tool	for	the	right	job”	… maybe	too	many
SIZE	PROBLEM	- SOLUTION
The	main	problem:	How	to	process	huge	and	
constantly	growing	amount	of	data	without	
downtimes.	
Technology	Solution:	Horizontally	scalable,	HA	and	
Multi-datacenter
Aerospike as	real-time	cookie	store
HDFS	+	Spark.
Programmatic	solution - analytics	approaches	as:
Sampling
Probabilistic	algorithms
Monitor as	much	as	possible
Graphite	is	our	best	friend	to	see	the	trends,	
anomalies,	performance	problems	and	predict	
potential	bottlenecks	in	advance.
DevOps
SIZE
Infrastructure
1300	bare	metal	servers	across	6	DCs (mostly HPE	ProLiant	DL360s	and	DL380s)
Private	Cloud
2400	OpenStack VMs
1050	Hyper-V VMs
OS	Distribution
Physical	servers:	50/50	Windows/Linux	
(mainly	CentOS)
VMs:	80%	Linux	/	20%	Windows
DATA	CENTERS	- GEO
Bare	metal
1	– Vilnius
3	– Copenhagen
1	– New	York
1	– Hong	Kong
AWS
São	Paulo
NETWORK
NIANET INT
10G	(DK)
TELIA INT
5G	(DK)
itxasr01
2.83G
2.94G
3.69G
2.58G
Netmon
Details
Dark	Fiber CWDM
PROVISIONING
Physical	servers:	Foreman up	to	the	OS,	
then	Puppet takes	over.	
VMs:	Puppet is	the	officially	supported	
method.
Terraform to	launch	VMs.
OS	images	are	standardized
puppet	- same	configuration	for	every	
server	
(NTP,	firewall	settings,	etc).
OPENSTACK	&	CEPH
Multi-region	self	service	for
Virtual	machines,	~2400	VMs
Network,	storage,	full	automation,	
integration	with	CI		&	CD	
Separate environments	for
Development		
Pre-production	(staging)	
Production
DATA	STORAGE
HDFS
ETL	cluster:
64	workers
HDFS:	2.5	PB
Memory:	9.84	TB
VCores:	2814
DMP	cluster:
40	workers
HDFS:	1.1	PB
Memory:	7.42	TB
VCores:	1548
HIVE	VS	SPARK
0
10
20
30
40
50
60
70
80
90
100
110
15	minutes 1	hour	15	minutes 4	hours	 1	day
Time		to	aggregate	one	day	in	minutes
Aggregation	chunk	size
ONE	DAY	AGGREGATION
Hive Spark	1.5 Spark	2.0
Hive
Spark	30x	faster
VERTICA
Scalable	Column	Store	- Great	for	complex	
multi-filter	queries
Total	data	volume	~250TB
Average	 query	latency	0.541	sec
Biggest	tables	100B	rows
30-40k queries	per	hour
Rack	aware	for	HA
Datacenter	aware	for	even	more	
Horizontally	scalable
30%	growth per	year	~4	nodes	to	cluster
SELECT avg(price) FROM tickstore
WHERE symbol = ‘AAPL”
AND date = ‘5/06/09’
Before	Vertica
We	used	to	use	MS	BI	stack- MS	SQL+MS	OLAP.	
“It	was	a	nightmare	to	develop	and	support	
the	solution,	especially	scaling	out”
Now	zero	complaints	
5/05/09
5/06/09
5/05/09
5/06/09
143.74
143.75
37.03
37.13
Column	Store	- Reads	3	columns
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
NYSE	
NYSE	
NYSE	
NQDS	
AAPL
AAPL
BBY
BBY
AEROSPIKE
Row	store	– Great	for	primary	key	queries
Total	data	volume	~15TB
Average		query	latency	- 0.9	millisecond
Biggest	set	- 8B	rows
~1	million	queries	per	second
Rack	aware	for	HA	and	DC	redundancy
Horizontally	scalable
55%	growth	each	~4	nodes	to	cluster	per	year
Before	Aerospike
Constantly	fighting	with	Cassandra
Java	GC	problems
Rebalancing	problems
Defrag/compaction	problems
SELECT device, geo, segments FROM
cookie_store
WHERE cookieId = 826701826
KAFKA/CONFLUENT
1	million messages	per	second	or	60	
billion	messages	a	day.	
Cluster	of	24	servers
225	TB	of	storage.	
3	remote	clusters	that	mirror
Moving	to	Confluent soon
UNCLE	PETE’S	ADVICE
DATA	FLOW
Data is	
the	blood
Use	Kafka	&	Confluent
Kafka is	the	
circulatory	system
Confluent is	a	
healthy	lifestyle
KAFKA’S	IMPACT	ON	DevOps
Kafka	as	a	data	pipeline	
Before	Kafka	each	team	would	
build	custom	pipelines	to	
share	data.
It	would	take	up	to	3	months	
to	implement.	
After	switching	to	Kafka	as	our	
data	pipeline,	any	team	can	
tap	into	the	stream and	use	
the	data.
DATA	STORAGE	AND	PROCESSING
SAVE	YOURSELF	SOME	PAIN
Use	Vertica as	your	
column	store	
Þ
Scale,	HA	- complex	
queries
Replace	Hive	with	
Spark
Þ
At	least	20x	faster	on	
same	hardware
Parquet and	HDFS	as	
your	raw	data	store
Þ
30%	less	space
Use	Aerospike as	
your	row	store	
Þ
Speed,	Scale,	HA	-
primary	key
TECHNOLOGY	TO	WATCH
Persistent	RAM
Graphics	Processing	Units	(GPUs)
Deep	learningColumnar	databases
OPEN	SOURCE	&	ENTERPRISE
Enterprise	is	value	for	money
Support
Training
Consulting
Contractual	obligation	- A	neck	to	choke
Open	source	!=	Free
Modern	escrow
Fork	the	source	repository
Build it	yourself
Survive on	your	own
VISIT	YOUR	VENDOR
Hip	startup or	stable
enterprise
Vision
Financial stability
Business	ethics
Good	Girls/Guys	or	…
Use	Continuous Integration	&	Delivery
We	use:	TeamCity	&	Rundeck
Build	in	blue/green deployment	&	Feature	
toggling
Develop	a	DevOps	culture – think	“how	will	it	
run	in	production”
Hire	good	people	– not	entitled	princesses
Use	an	Agile process	- Agile	is:
A	frame	of	mind
A	flexible	technique to	manage	a	project
Allows	new	information	to	alter	outcome
PROCESS
AGILE	IS	NOT
Not	a	club,	social	group	or	a	Cult
Not	an	excuse	for	being	disorganized,	
bored	or	having	a	short	attention	span
MURPHY’S	LAW
Anything	you	haven’t	tested	
will	break
Entropy – 3rd law	of	
thermodynamics
WISDOM
Solution	should	be	as	simple	as	possible,	
but	no	simpler
A	Hypothesis is	a	statement	whose	truth is	
temporarily	assumed,	but	whose	meaning
must	be	beyond	all	doubt
PARALLELISM	AT	EVERY	STEP
Multiple	datacenters
Multiple	power	sources
Multiple	network	paths
Multiple	machines
Multiple	cores
Multi-channel	Bus
Multiple	IRQs
Multiple	processes
Multiple	threads
FINALLY
No	solution	is	successful	unless	it	delivers	
RUNNING	CODE
Regardless	of	methodology,	team	spirit,	coding	dogma	
or	popular	fashion
@AdformInsider

More Related Content

Viewers also liked

Nagoya.R #12 Rprofile作成のススメ
Nagoya.R #12 Rprofile作成のススメNagoya.R #12 Rprofile作成のススメ
Nagoya.R #12 Rprofile作成のススメ
Yusaku Kawaguchi
 
Fail over fail_back
Fail over fail_backFail over fail_back
Fail over fail_back
PostgreSQL Experts, Inc.
 
Gedetineerden blijven er lustig op los bellen
Gedetineerden blijven er lustig op los bellenGedetineerden blijven er lustig op los bellen
Gedetineerden blijven er lustig op los bellen
Thierry Debels
 
2 bhk Bedroom Builder Floor for Rent in Gurgaon
2 bhk  Bedroom Builder Floor for Rent in Gurgaon2 bhk  Bedroom Builder Floor for Rent in Gurgaon
2 bhk Bedroom Builder Floor for Rent in Gurgaon
Vivek Bhaskar
 
Ciberinteligencia2
Ciberinteligencia2Ciberinteligencia2
E2D3グラフの投稿ハンズオン
E2D3グラフの投稿ハンズオンE2D3グラフの投稿ハンズオン
E2D3グラフの投稿ハンズオン
Junichi Watanuki
 
Bilangan Peroksida dan Bilangan TBA
Bilangan Peroksida dan Bilangan TBABilangan Peroksida dan Bilangan TBA
Bilangan Peroksida dan Bilangan TBA
Yokhebed Fransisca
 
#ChangeAgents, Experiments, & Expertise in Our Exponential Era - David Bray
#ChangeAgents, Experiments, & Expertise in Our Exponential Era - David Bray #ChangeAgents, Experiments, & Expertise in Our Exponential Era - David Bray
#ChangeAgents, Experiments, & Expertise in Our Exponential Era - David Bray
scoopnewsgroup
 
Faster Smarter Decision Cycles: The Key to Winning in the New Normal
Faster Smarter Decision Cycles: The Key to Winning in the New NormalFaster Smarter Decision Cycles: The Key to Winning in the New Normal
Faster Smarter Decision Cycles: The Key to Winning in the New Normal
Joyce_FastTime
 
Accélérateurs - Valeurs Actuelles
Accélérateurs - Valeurs ActuellesAccélérateurs - Valeurs Actuelles
Accélérateurs - Valeurs Actuelles
Nicolas Dufourcq
 
InsideSales.com - The Best Practices For Lead Response Management
InsideSales.com - The Best Practices For Lead Response ManagementInsideSales.com - The Best Practices For Lead Response Management
InsideSales.com - The Best Practices For Lead Response Management
InsideSales.com
 
Leona Chin Profile 2017
Leona Chin Profile 2017Leona Chin Profile 2017
Leona Chin Profile 2017
Leona Chin
 
Trump planta cara a Basilea
Trump planta cara a BasileaTrump planta cara a Basilea
Trump planta cara a Basilea
Gabriel Holand
 
Emploi et salaires du design interactif : enquête 2016
Emploi et salaires du design interactif : enquête 2016Emploi et salaires du design interactif : enquête 2016
Emploi et salaires du design interactif : enquête 2016
designers interactifs
 
Where next for corrugated cardboard packaging in Europe ?
Where next for corrugated cardboard packaging in Europe ?Where next for corrugated cardboard packaging in Europe ?
Where next for corrugated cardboard packaging in Europe ?
Katie Ryan
 
Machine learning by example
Machine learning by exampleMachine learning by example
Machine learning by example
SoftwareMill
 
Passme CoP meeting customised interiors A. Castro Nov.2016
Passme CoP meeting customised interiors A. Castro Nov.2016Passme CoP meeting customised interiors A. Castro Nov.2016
Passme CoP meeting customised interiors A. Castro Nov.2016
PASSME Community
 

Viewers also liked (18)

Nagoya.R #12 Rprofile作成のススメ
Nagoya.R #12 Rprofile作成のススメNagoya.R #12 Rprofile作成のススメ
Nagoya.R #12 Rprofile作成のススメ
 
Fail over fail_back
Fail over fail_backFail over fail_back
Fail over fail_back
 
Gedetineerden blijven er lustig op los bellen
Gedetineerden blijven er lustig op los bellenGedetineerden blijven er lustig op los bellen
Gedetineerden blijven er lustig op los bellen
 
2 bhk Bedroom Builder Floor for Rent in Gurgaon
2 bhk  Bedroom Builder Floor for Rent in Gurgaon2 bhk  Bedroom Builder Floor for Rent in Gurgaon
2 bhk Bedroom Builder Floor for Rent in Gurgaon
 
Ciberinteligencia2
Ciberinteligencia2Ciberinteligencia2
Ciberinteligencia2
 
E2D3グラフの投稿ハンズオン
E2D3グラフの投稿ハンズオンE2D3グラフの投稿ハンズオン
E2D3グラフの投稿ハンズオン
 
Bilangan Peroksida dan Bilangan TBA
Bilangan Peroksida dan Bilangan TBABilangan Peroksida dan Bilangan TBA
Bilangan Peroksida dan Bilangan TBA
 
#ChangeAgents, Experiments, & Expertise in Our Exponential Era - David Bray
#ChangeAgents, Experiments, & Expertise in Our Exponential Era - David Bray #ChangeAgents, Experiments, & Expertise in Our Exponential Era - David Bray
#ChangeAgents, Experiments, & Expertise in Our Exponential Era - David Bray
 
Faster Smarter Decision Cycles: The Key to Winning in the New Normal
Faster Smarter Decision Cycles: The Key to Winning in the New NormalFaster Smarter Decision Cycles: The Key to Winning in the New Normal
Faster Smarter Decision Cycles: The Key to Winning in the New Normal
 
Accélérateurs - Valeurs Actuelles
Accélérateurs - Valeurs ActuellesAccélérateurs - Valeurs Actuelles
Accélérateurs - Valeurs Actuelles
 
InsideSales.com - The Best Practices For Lead Response Management
InsideSales.com - The Best Practices For Lead Response ManagementInsideSales.com - The Best Practices For Lead Response Management
InsideSales.com - The Best Practices For Lead Response Management
 
Leona Chin Profile 2017
Leona Chin Profile 2017Leona Chin Profile 2017
Leona Chin Profile 2017
 
Trump planta cara a Basilea
Trump planta cara a BasileaTrump planta cara a Basilea
Trump planta cara a Basilea
 
Emploi et salaires du design interactif : enquête 2016
Emploi et salaires du design interactif : enquête 2016Emploi et salaires du design interactif : enquête 2016
Emploi et salaires du design interactif : enquête 2016
 
Where next for corrugated cardboard packaging in Europe ?
Where next for corrugated cardboard packaging in Europe ?Where next for corrugated cardboard packaging in Europe ?
Where next for corrugated cardboard packaging in Europe ?
 
Creative commons de
Creative commons deCreative commons de
Creative commons de
 
Machine learning by example
Machine learning by exampleMachine learning by example
Machine learning by example
 
Passme CoP meeting customised interiors A. Castro Nov.2016
Passme CoP meeting customised interiors A. Castro Nov.2016Passme CoP meeting customised interiors A. Castro Nov.2016
Passme CoP meeting customised interiors A. Castro Nov.2016
 

Similar to Achieving High Load in Advertising Technology

Dell Technologies - Company and Portfolio Introduction in 20 Minutes
Dell Technologies - Company and Portfolio Introduction in 20 MinutesDell Technologies - Company and Portfolio Introduction in 20 Minutes
Dell Technologies - Company and Portfolio Introduction in 20 Minutes
Smarter.World
 
Winnipeg Technology Innovation Day
Winnipeg Technology Innovation Day Winnipeg Technology Innovation Day
Winnipeg Technology Innovation Day
Acrodex
 
Maximizing Big Data ROI via Best of Breed Technology Patterns and Practices -...
Maximizing Big Data ROI via Best of Breed Technology Patterns and Practices -...Maximizing Big Data ROI via Best of Breed Technology Patterns and Practices -...
Maximizing Big Data ROI via Best of Breed Technology Patterns and Practices -...
DigiMarCon - Digital Marketing, Media and Advertising Conferences & Exhibitions
 
Business Together as a Service global event, April 23, 2013
Business Together as a Service global event, April 23, 2013Business Together as a Service global event, April 23, 2013
Business Together as a Service global event, April 23, 2013
Orange Business Services
 
Leadership Exchange Houston Shared Presentations
Leadership Exchange Houston Shared PresentationsLeadership Exchange Houston Shared Presentations
Leadership Exchange Houston Shared Presentations
StephanieCaldwell9
 
seamless data.pptx jhuuhuhuhuhhuhjnjokojijoijij
seamless data.pptx    jhuuhuhuhuhhuhjnjokojijoijijseamless data.pptx    jhuuhuhuhuhhuhjnjokojijoijij
seamless data.pptx jhuuhuhuhuhhuhjnjokojijoijij
salmanfrel434
 
Keynot Cloud Expo Intel
Keynot Cloud Expo IntelKeynot Cloud Expo Intel
Keynot Cloud Expo Intel
Quentin Adam
 
JBoss Enterprise Middleware: Beneficios y propuesta de valor - Consuelo Sánch...
JBoss Enterprise Middleware: Beneficios y propuesta de valor - Consuelo Sánch...JBoss Enterprise Middleware: Beneficios y propuesta de valor - Consuelo Sánch...
JBoss Enterprise Middleware: Beneficios y propuesta de valor - Consuelo Sánch...
Nextel S.A.
 
Top 5 In-demand technologies to Learn in 2020
Top 5 In-demand technologies to Learn in 2020Top 5 In-demand technologies to Learn in 2020
Top 5 In-demand technologies to Learn in 2020
Intellipaat
 
Top 5 In-demand Technologies to Learn in 2020
Top 5 In-demand Technologies to Learn in 2020Top 5 In-demand Technologies to Learn in 2020
Top 5 In-demand Technologies to Learn in 2020
Intellipaat
 
Practical advice to build a data driven company
Practical advice to build a data driven companyPractical advice to build a data driven company
Practical advice to build a data driven company
DataWorks Summit/Hadoop Summit
 
Compuware Company Profile 2020
Compuware Company Profile 2020Compuware Company Profile 2020
Compuware Company Profile 2020
Nicole Yu
 
Digital transformation with microsoft data and ai
Digital transformation with microsoft data and ai Digital transformation with microsoft data and ai
Digital transformation with microsoft data and ai
MichaelRoenker
 
How to develop a multi cloud strategy to accelerate digital transformation - ...
How to develop a multi cloud strategy to accelerate digital transformation - ...How to develop a multi cloud strategy to accelerate digital transformation - ...
How to develop a multi cloud strategy to accelerate digital transformation - ...
Senaka Ariyasinghe
 
Plex Sales Handout - Mitchell Plastics v5_final
Plex Sales Handout - Mitchell Plastics v5_finalPlex Sales Handout - Mitchell Plastics v5_final
Plex Sales Handout - Mitchell Plastics v5_finalDavid Klotz
 
Presenting SmartSource
Presenting SmartSourcePresenting SmartSource
Presenting SmartSource
Robert Lind
 
ASCII Austin Keynote by Intel SMB MSP Marketing Director, Eric Townsend keyno...
ASCII Austin Keynote by Intel SMB MSP Marketing Director, Eric Townsend keyno...ASCII Austin Keynote by Intel SMB MSP Marketing Director, Eric Townsend keyno...
ASCII Austin Keynote by Intel SMB MSP Marketing Director, Eric Townsend keyno...
IntelMSP
 
VoiceCon Orlando 2010 Slidecast 2
VoiceCon Orlando 2010 Slidecast 2VoiceCon Orlando 2010 Slidecast 2
VoiceCon Orlando 2010 Slidecast 2ekrapf
 
Dell EMC - proposta accordo quadro CRUI
Dell EMC - proposta accordo quadro CRUIDell EMC - proposta accordo quadro CRUI
Dell EMC - proposta accordo quadro CRUI
Jürgen Ambrosi
 
Citihub IDC Event 2009 Cloud Mark Ellis
Citihub IDC Event 2009 Cloud Mark EllisCitihub IDC Event 2009 Cloud Mark Ellis
Citihub IDC Event 2009 Cloud Mark Ellis
markellis
 

Similar to Achieving High Load in Advertising Technology (20)

Dell Technologies - Company and Portfolio Introduction in 20 Minutes
Dell Technologies - Company and Portfolio Introduction in 20 MinutesDell Technologies - Company and Portfolio Introduction in 20 Minutes
Dell Technologies - Company and Portfolio Introduction in 20 Minutes
 
Winnipeg Technology Innovation Day
Winnipeg Technology Innovation Day Winnipeg Technology Innovation Day
Winnipeg Technology Innovation Day
 
Maximizing Big Data ROI via Best of Breed Technology Patterns and Practices -...
Maximizing Big Data ROI via Best of Breed Technology Patterns and Practices -...Maximizing Big Data ROI via Best of Breed Technology Patterns and Practices -...
Maximizing Big Data ROI via Best of Breed Technology Patterns and Practices -...
 
Business Together as a Service global event, April 23, 2013
Business Together as a Service global event, April 23, 2013Business Together as a Service global event, April 23, 2013
Business Together as a Service global event, April 23, 2013
 
Leadership Exchange Houston Shared Presentations
Leadership Exchange Houston Shared PresentationsLeadership Exchange Houston Shared Presentations
Leadership Exchange Houston Shared Presentations
 
seamless data.pptx jhuuhuhuhuhhuhjnjokojijoijij
seamless data.pptx    jhuuhuhuhuhhuhjnjokojijoijijseamless data.pptx    jhuuhuhuhuhhuhjnjokojijoijij
seamless data.pptx jhuuhuhuhuhhuhjnjokojijoijij
 
Keynot Cloud Expo Intel
Keynot Cloud Expo IntelKeynot Cloud Expo Intel
Keynot Cloud Expo Intel
 
JBoss Enterprise Middleware: Beneficios y propuesta de valor - Consuelo Sánch...
JBoss Enterprise Middleware: Beneficios y propuesta de valor - Consuelo Sánch...JBoss Enterprise Middleware: Beneficios y propuesta de valor - Consuelo Sánch...
JBoss Enterprise Middleware: Beneficios y propuesta de valor - Consuelo Sánch...
 
Top 5 In-demand technologies to Learn in 2020
Top 5 In-demand technologies to Learn in 2020Top 5 In-demand technologies to Learn in 2020
Top 5 In-demand technologies to Learn in 2020
 
Top 5 In-demand Technologies to Learn in 2020
Top 5 In-demand Technologies to Learn in 2020Top 5 In-demand Technologies to Learn in 2020
Top 5 In-demand Technologies to Learn in 2020
 
Practical advice to build a data driven company
Practical advice to build a data driven companyPractical advice to build a data driven company
Practical advice to build a data driven company
 
Compuware Company Profile 2020
Compuware Company Profile 2020Compuware Company Profile 2020
Compuware Company Profile 2020
 
Digital transformation with microsoft data and ai
Digital transformation with microsoft data and ai Digital transformation with microsoft data and ai
Digital transformation with microsoft data and ai
 
How to develop a multi cloud strategy to accelerate digital transformation - ...
How to develop a multi cloud strategy to accelerate digital transformation - ...How to develop a multi cloud strategy to accelerate digital transformation - ...
How to develop a multi cloud strategy to accelerate digital transformation - ...
 
Plex Sales Handout - Mitchell Plastics v5_final
Plex Sales Handout - Mitchell Plastics v5_finalPlex Sales Handout - Mitchell Plastics v5_final
Plex Sales Handout - Mitchell Plastics v5_final
 
Presenting SmartSource
Presenting SmartSourcePresenting SmartSource
Presenting SmartSource
 
ASCII Austin Keynote by Intel SMB MSP Marketing Director, Eric Townsend keyno...
ASCII Austin Keynote by Intel SMB MSP Marketing Director, Eric Townsend keyno...ASCII Austin Keynote by Intel SMB MSP Marketing Director, Eric Townsend keyno...
ASCII Austin Keynote by Intel SMB MSP Marketing Director, Eric Townsend keyno...
 
VoiceCon Orlando 2010 Slidecast 2
VoiceCon Orlando 2010 Slidecast 2VoiceCon Orlando 2010 Slidecast 2
VoiceCon Orlando 2010 Slidecast 2
 
Dell EMC - proposta accordo quadro CRUI
Dell EMC - proposta accordo quadro CRUIDell EMC - proposta accordo quadro CRUI
Dell EMC - proposta accordo quadro CRUI
 
Citihub IDC Event 2009 Cloud Mark Ellis
Citihub IDC Event 2009 Cloud Mark EllisCitihub IDC Event 2009 Cloud Mark Ellis
Citihub IDC Event 2009 Cloud Mark Ellis
 

More from Peter Milne

Principles of High Load - Vilnius January 2015
Principles of High Load - Vilnius January 2015Principles of High Load - Vilnius January 2015
Principles of High Load - Vilnius January 2015
Peter Milne
 
Aerospike Architecture
Aerospike ArchitectureAerospike Architecture
Aerospike Architecture
Peter Milne
 
Recommendation engine using Aerospike and/OR MongoDB
Recommendation engine using Aerospike and/OR MongoDBRecommendation engine using Aerospike and/OR MongoDB
Recommendation engine using Aerospike and/OR MongoDB
Peter Milne
 
Real Time Analytics
Real Time AnalyticsReal Time Analytics
Real Time Analytics
Peter Milne
 
Generating NoSQL from SQL
Generating NoSQL from SQLGenerating NoSQL from SQL
Generating NoSQL from SQL
Peter Milne
 
Aerospike Architecture
Aerospike ArchitectureAerospike Architecture
Aerospike Architecture
Peter Milne
 

More from Peter Milne (6)

Principles of High Load - Vilnius January 2015
Principles of High Load - Vilnius January 2015Principles of High Load - Vilnius January 2015
Principles of High Load - Vilnius January 2015
 
Aerospike Architecture
Aerospike ArchitectureAerospike Architecture
Aerospike Architecture
 
Recommendation engine using Aerospike and/OR MongoDB
Recommendation engine using Aerospike and/OR MongoDBRecommendation engine using Aerospike and/OR MongoDB
Recommendation engine using Aerospike and/OR MongoDB
 
Real Time Analytics
Real Time AnalyticsReal Time Analytics
Real Time Analytics
 
Generating NoSQL from SQL
Generating NoSQL from SQLGenerating NoSQL from SQL
Generating NoSQL from SQL
 
Aerospike Architecture
Aerospike ArchitectureAerospike Architecture
Aerospike Architecture
 

Recently uploaded

Into the Box 2024 - Keynote Day 2 Slides.pdf
Into the Box 2024 - Keynote Day 2 Slides.pdfInto the Box 2024 - Keynote Day 2 Slides.pdf
Into the Box 2024 - Keynote Day 2 Slides.pdf
Ortus Solutions, Corp
 
A Comprehensive Look at Generative AI in Retail App Testing.pdf
A Comprehensive Look at Generative AI in Retail App Testing.pdfA Comprehensive Look at Generative AI in Retail App Testing.pdf
A Comprehensive Look at Generative AI in Retail App Testing.pdf
kalichargn70th171
 
Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...
Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...
Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...
Shahin Sheidaei
 
Vitthal Shirke Microservices Resume Montevideo
Vitthal Shirke Microservices Resume MontevideoVitthal Shirke Microservices Resume Montevideo
Vitthal Shirke Microservices Resume Montevideo
Vitthal Shirke
 
Enhancing Project Management Efficiency_ Leveraging AI Tools like ChatGPT.pdf
Enhancing Project Management Efficiency_ Leveraging AI Tools like ChatGPT.pdfEnhancing Project Management Efficiency_ Leveraging AI Tools like ChatGPT.pdf
Enhancing Project Management Efficiency_ Leveraging AI Tools like ChatGPT.pdf
Jay Das
 
Orion Context Broker introduction 20240604
Orion Context Broker introduction 20240604Orion Context Broker introduction 20240604
Orion Context Broker introduction 20240604
Fermin Galan
 
Globus Compute wth IRI Workflows - GlobusWorld 2024
Globus Compute wth IRI Workflows - GlobusWorld 2024Globus Compute wth IRI Workflows - GlobusWorld 2024
Globus Compute wth IRI Workflows - GlobusWorld 2024
Globus
 
Graphic Design Crash Course for beginners
Graphic Design Crash Course for beginnersGraphic Design Crash Course for beginners
Graphic Design Crash Course for beginners
e20449
 
Dominate Social Media with TubeTrivia AI’s Addictive Quiz Videos.pdf
Dominate Social Media with TubeTrivia AI’s Addictive Quiz Videos.pdfDominate Social Media with TubeTrivia AI’s Addictive Quiz Videos.pdf
Dominate Social Media with TubeTrivia AI’s Addictive Quiz Videos.pdf
AMB-Review
 
Cracking the code review at SpringIO 2024
Cracking the code review at SpringIO 2024Cracking the code review at SpringIO 2024
Cracking the code review at SpringIO 2024
Paco van Beckhoven
 
Enterprise Resource Planning System in Telangana
Enterprise Resource Planning System in TelanganaEnterprise Resource Planning System in Telangana
Enterprise Resource Planning System in Telangana
NYGGS Automation Suite
 
In 2015, I used to write extensions for Joomla, WordPress, phpBB3, etc and I ...
In 2015, I used to write extensions for Joomla, WordPress, phpBB3, etc and I ...In 2015, I used to write extensions for Joomla, WordPress, phpBB3, etc and I ...
In 2015, I used to write extensions for Joomla, WordPress, phpBB3, etc and I ...
Juraj Vysvader
 
AI Pilot Review: The World’s First Virtual Assistant Marketing Suite
AI Pilot Review: The World’s First Virtual Assistant Marketing SuiteAI Pilot Review: The World’s First Virtual Assistant Marketing Suite
AI Pilot Review: The World’s First Virtual Assistant Marketing Suite
Google
 
Prosigns: Transforming Business with Tailored Technology Solutions
Prosigns: Transforming Business with Tailored Technology SolutionsProsigns: Transforming Business with Tailored Technology Solutions
Prosigns: Transforming Business with Tailored Technology Solutions
Prosigns
 
TROUBLESHOOTING 9 TYPES OF OUTOFMEMORYERROR
TROUBLESHOOTING 9 TYPES OF OUTOFMEMORYERRORTROUBLESHOOTING 9 TYPES OF OUTOFMEMORYERROR
TROUBLESHOOTING 9 TYPES OF OUTOFMEMORYERROR
Tier1 app
 
RISE with SAP and Journey to the Intelligent Enterprise
RISE with SAP and Journey to the Intelligent EnterpriseRISE with SAP and Journey to the Intelligent Enterprise
RISE with SAP and Journey to the Intelligent Enterprise
Srikant77
 
2024 RoOUG Security model for the cloud.pptx
2024 RoOUG Security model for the cloud.pptx2024 RoOUG Security model for the cloud.pptx
2024 RoOUG Security model for the cloud.pptx
Georgi Kodinov
 
Navigating the Metaverse: A Journey into Virtual Evolution"
Navigating the Metaverse: A Journey into Virtual Evolution"Navigating the Metaverse: A Journey into Virtual Evolution"
Navigating the Metaverse: A Journey into Virtual Evolution"
Donna Lenk
 
Enhancing Research Orchestration Capabilities at ORNL.pdf
Enhancing Research Orchestration Capabilities at ORNL.pdfEnhancing Research Orchestration Capabilities at ORNL.pdf
Enhancing Research Orchestration Capabilities at ORNL.pdf
Globus
 
Climate Science Flows: Enabling Petabyte-Scale Climate Analysis with the Eart...
Climate Science Flows: Enabling Petabyte-Scale Climate Analysis with the Eart...Climate Science Flows: Enabling Petabyte-Scale Climate Analysis with the Eart...
Climate Science Flows: Enabling Petabyte-Scale Climate Analysis with the Eart...
Globus
 

Recently uploaded (20)

Into the Box 2024 - Keynote Day 2 Slides.pdf
Into the Box 2024 - Keynote Day 2 Slides.pdfInto the Box 2024 - Keynote Day 2 Slides.pdf
Into the Box 2024 - Keynote Day 2 Slides.pdf
 
A Comprehensive Look at Generative AI in Retail App Testing.pdf
A Comprehensive Look at Generative AI in Retail App Testing.pdfA Comprehensive Look at Generative AI in Retail App Testing.pdf
A Comprehensive Look at Generative AI in Retail App Testing.pdf
 
Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...
Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...
Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...
 
Vitthal Shirke Microservices Resume Montevideo
Vitthal Shirke Microservices Resume MontevideoVitthal Shirke Microservices Resume Montevideo
Vitthal Shirke Microservices Resume Montevideo
 
Enhancing Project Management Efficiency_ Leveraging AI Tools like ChatGPT.pdf
Enhancing Project Management Efficiency_ Leveraging AI Tools like ChatGPT.pdfEnhancing Project Management Efficiency_ Leveraging AI Tools like ChatGPT.pdf
Enhancing Project Management Efficiency_ Leveraging AI Tools like ChatGPT.pdf
 
Orion Context Broker introduction 20240604
Orion Context Broker introduction 20240604Orion Context Broker introduction 20240604
Orion Context Broker introduction 20240604
 
Globus Compute wth IRI Workflows - GlobusWorld 2024
Globus Compute wth IRI Workflows - GlobusWorld 2024Globus Compute wth IRI Workflows - GlobusWorld 2024
Globus Compute wth IRI Workflows - GlobusWorld 2024
 
Graphic Design Crash Course for beginners
Graphic Design Crash Course for beginnersGraphic Design Crash Course for beginners
Graphic Design Crash Course for beginners
 
Dominate Social Media with TubeTrivia AI’s Addictive Quiz Videos.pdf
Dominate Social Media with TubeTrivia AI’s Addictive Quiz Videos.pdfDominate Social Media with TubeTrivia AI’s Addictive Quiz Videos.pdf
Dominate Social Media with TubeTrivia AI’s Addictive Quiz Videos.pdf
 
Cracking the code review at SpringIO 2024
Cracking the code review at SpringIO 2024Cracking the code review at SpringIO 2024
Cracking the code review at SpringIO 2024
 
Enterprise Resource Planning System in Telangana
Enterprise Resource Planning System in TelanganaEnterprise Resource Planning System in Telangana
Enterprise Resource Planning System in Telangana
 
In 2015, I used to write extensions for Joomla, WordPress, phpBB3, etc and I ...
In 2015, I used to write extensions for Joomla, WordPress, phpBB3, etc and I ...In 2015, I used to write extensions for Joomla, WordPress, phpBB3, etc and I ...
In 2015, I used to write extensions for Joomla, WordPress, phpBB3, etc and I ...
 
AI Pilot Review: The World’s First Virtual Assistant Marketing Suite
AI Pilot Review: The World’s First Virtual Assistant Marketing SuiteAI Pilot Review: The World’s First Virtual Assistant Marketing Suite
AI Pilot Review: The World’s First Virtual Assistant Marketing Suite
 
Prosigns: Transforming Business with Tailored Technology Solutions
Prosigns: Transforming Business with Tailored Technology SolutionsProsigns: Transforming Business with Tailored Technology Solutions
Prosigns: Transforming Business with Tailored Technology Solutions
 
TROUBLESHOOTING 9 TYPES OF OUTOFMEMORYERROR
TROUBLESHOOTING 9 TYPES OF OUTOFMEMORYERRORTROUBLESHOOTING 9 TYPES OF OUTOFMEMORYERROR
TROUBLESHOOTING 9 TYPES OF OUTOFMEMORYERROR
 
RISE with SAP and Journey to the Intelligent Enterprise
RISE with SAP and Journey to the Intelligent EnterpriseRISE with SAP and Journey to the Intelligent Enterprise
RISE with SAP and Journey to the Intelligent Enterprise
 
2024 RoOUG Security model for the cloud.pptx
2024 RoOUG Security model for the cloud.pptx2024 RoOUG Security model for the cloud.pptx
2024 RoOUG Security model for the cloud.pptx
 
Navigating the Metaverse: A Journey into Virtual Evolution"
Navigating the Metaverse: A Journey into Virtual Evolution"Navigating the Metaverse: A Journey into Virtual Evolution"
Navigating the Metaverse: A Journey into Virtual Evolution"
 
Enhancing Research Orchestration Capabilities at ORNL.pdf
Enhancing Research Orchestration Capabilities at ORNL.pdfEnhancing Research Orchestration Capabilities at ORNL.pdf
Enhancing Research Orchestration Capabilities at ORNL.pdf
 
Climate Science Flows: Enabling Petabyte-Scale Climate Analysis with the Eart...
Climate Science Flows: Enabling Petabyte-Scale Climate Analysis with the Eart...Climate Science Flows: Enabling Petabyte-Scale Climate Analysis with the Eart...
Climate Science Flows: Enabling Petabyte-Scale Climate Analysis with the Eart...
 

Achieving High Load in Advertising Technology