Slide 1

Slide 1 text

Accelerating an Ad-Tech Service with New Data Analyzing Platform CyberAgent, Inc Makoto Hasegawa

Slide 2

Slide 2 text

2 About me Makoto Hasegawa I am a cloud architect and leading the system admin team of CyberAgent Ad-Tech Business Division. I have been managing the OpenStack for about 2 year. Other than OpenStack, I also use cloud platforms such as AWS and GCP. Recently I’m most interested in how I can automate the cloud management by using various different tools. @makocchi https://facebook.com/makocchi0923

Slide 3

Slide 3 text

3 Agenda ɾAbout CyberAgent and adtech studio ɾAbout adtech studio infrastructures ɾAbout new data analyzing platform ɾAbout hardware for platform ɾFAQ

Slide 4

Slide 4 text

4 CyberAgent ? About CyberAgent and adtech studio About adtech studio infrastructures About new data analyzing platform About hardware for platform Accelerating an Ad-Tech Service with New Data Analyzing Platform

Slide 5

Slide 5 text

5 “To create the 21st Century’s leading company” CyberAgent is expanding its business in the field of Internet, A leading industry of the 21st Century.

Slide 6

Slide 6 text

6 About CyberAgent We are developing media business, Internet advertisement business, and game business etc.

Slide 7

Slide 7 text

7 About CyberAgent Internet advertisement business has been deployed in CyberAgent and 21 consolidated subsidiaries, its advertising billings is the number one in Japan. We sell internet ads (smartphone/PC) to advertisers. Internet advertisement business *Source: Joint survey of CyberAgent and Digital Infact of Seed Planning; Calculated based on our financial results for the period from Jan. to Dec. 2015 and the reference materials disclosed by each company.

Slide 8

Slide 8 text

8 About CyberAgent Game business, which was started from 2009, offers seven major titles and other several tens of smartphone games. Led by Cygames Inc., 11 subsidiaries specializing in game productions are developing and running smartphone games of various genres. Game business

Slide 9

Slide 9 text

9 About CyberAgent In 2004, A blog-based service “Ameba” has started. Since then our services changes its form to avatar service “Ameba Pigg”, to online community service to contents curation media, in accordance with the shift in the internet industry. In 2016, CyberAgent has entered the video business, which is attracting attention as a growth field. Media business

Slide 10

Slide 10 text

10 About CyberAgent Adtech Studio is a group of engineers at CyberAgent and its subsidiaries working to develop and operate the group’s advertising technology services. Established in October 2013 and comprising seven subsidiaries, it includes more than 250 engineers developing over 20 services. Adtech Studio

Slide 11

Slide 11 text

11 About CyberAgent Adtech Studio (Structure of Ad-Tech)

Slide 12

Slide 12 text

12 About CyberAgent Adtech Studio (FYI) If you want to know more details of Ad-Tech, please visit our website! https://www.cyberagent.co.jp/en/ir/personal/adtech/

Slide 13

Slide 13 text

13 Adtech studio infrastructures About CyberAgent and adtech studio About adtech studio infrastructures About new data analyzing platform About hardware for platform Accelerating an Ad-Tech Service with New Data Analyzing Platform

Slide 14

Slide 14 text

14 About adtech studio infrastructures We need …. ɾto accommodates a wide variety of subsidiaries and systems in a . ɾto be able to to changing policy and demand . ɾto deploy our infrastructure without the complex ɹconfiguration. ɾto raise when we want to deploy. ɹ(as a result, we are managing multi-vendor environment) ɾ . ɾ to support a variety of big data analysis. multi-tenant respond quickly as simple as possible the best hardware in each case redundant 10Gb networks robust platform

Slide 15

Slide 15 text

15 New data analyzing platform About CyberAgent and adtech studio About adtech studio infrastructures About new data analyzing platform About hardware for platform Accelerating an Ad-Tech Service with New Data Analyzing Platform

Slide 16

Slide 16 text

16 We need a robust platform for data analyzing We considered that what is required for deploying data analyzing platform. ɾmore CPU cores and more memories (but not a lot of disks) ɹ> We think that high-density servers are suitable in this case. ɾmore network bandwidth rather than 10Gb ɹ> 25Gb or 40Gb ? ɾmore efficient

Slide 17

Slide 17 text

17 platform platform platform platform Previously, our products had been analyzing in each. We thought that this is not optimal way. not efficient …

Slide 18

Slide 18 text

18 New Platform So we deployed new data analyzing platform and now we are accelerating an analyzing data. more efficient!

Slide 19

Slide 19 text

19 Hardware Accelerating an Ad-Tech Service with New Data Analyzing Platform About CyberAgent and adtech studio About adtech studio infrastructures About new data analyzing platform About hardware for platform

Slide 20

Slide 20 text

20 Hardware for platform High-density servers We chose Fusion Server X6800 and Fusion Server XH620 V3

Slide 21

Slide 21 text

21 Hardware for platform X6800 + XH620 V3 8 XH620 nodes in X6800 4U chassis (50% cut) 1 management UTP for 8 nodes (87.5% cut) 4 power supplies for 8 nodes (75% cut)

Slide 22

Slide 22 text

22 Hardware for platform 40Gb network We chose CE7850

Slide 23

Slide 23 text

23 Hardware for platform 40Gb network 32 x 40G QSFP+ port Managing CE7850 by iStack Can easy to version-up stacked switches

Slide 24

Slide 24 text

24 Hardware for platform Why Huawei? Huawei’s iStack was very nice for us. The response to some feature requests by tight cooperation with R&D was good. Always quick response when we needed support.

Slide 25

Slide 25 text

25 F A Q If you have any questions or anything, please do not hesitate to contact me! [email protected]

Slide 26

Slide 26 text

Copyright©2015 Huawei Technologies Co., Ltd. All Rights Reserved. The information in this document may contain predictive statements including, without limitation, statements regarding the future financial and operating results, future product portfolio, new technology, etc. There are a number of factors that could cause actual results and developments to differ materially from those expressed or implied in the predictive statements. Therefore, such information is provided for reference purpose only and constitutes neither an offer nor an acceptance. Huawei may change the information at any time without notice. Thank You