Monolithic 3D Inc., the Next Generation 3D-IC Company
  • Home
  • Technology
    • Technology
    • Papers, Presentations and Patents
    • Overview >
      • Background
      • Why Monolithic 3D?
      • Paths to Monolithic 3D
      • Applications
    • Ion-Cut: The Building Block
    • Monolithic 3D Logic >
      • RCAT
      • HKMG
      • Laser Annealing
      • RCJLT
      • 3D Embedded RAM
      • 3D Gate Array
      • FPGA
      • Ultra Large Integration - Redundancy and Repair
    • Monolithic 3D Memory >
      • 3D DRAM
      • 3D Resistive Memories
      • 3D Flash
    • Monolithic 3D Electro-Optics >
      • 3D Image Sensors
      • 3D Micro-Displays
  • 3D-IC Edge
    • 3D-IC Edge
  • News & Events
    • News & Events
    • S3S15 Game Change 2.0 Video/P
    • Webcast
    • Webinar
    • Press Releases
    • In the News
    • Upcoming Events
  • About Us
    • About Us
    • History
    • Team
    • Careers
    • Contact Us
  • Blog
  • Simulators

How Korea Became the Hub of the Memory Industry

11/20/2011

7 Comments

 
As you'd know, Korean companies such as Samsung and Hynix contribute 50-60% of the world's memory revenues. In today's blog post, we’ll look at reasons and strategies behind this dominance...

Like many researchers, I occasionally travel outside the country to attend conferences and give talks. I was in Korea for one such visit last week... It was my first time there, and I found it fascinating that such a small country contributes more than 50% of the world's memory output. Yes, its true, check out Figure 1. As the figure indicates, this Korean dominance was not sudden... it happened over a period of ten years between 1994 and 2004. Before that, Japanese chip makers (1984-1994) and American chip makers (before 1984) dominated the memory market. How did Korea achieve this phenomenal growth? Was it because of government policy, technical breakthroughs or economic conditions? Let's take a look...

Picture
Figure 1: Geographical distribution of the world's memory revenue. Includes DRAM and NAND flash memory.

Any study of Korean industry begins with an introduction to the chaebol business system. Chaebol are essentially a conglomerate of businesses, often owned by a single family. Through aggressive governmental support and finance, some chaebol have become well-known international brand names, such as Samsung, Hyundai and LG. For example, the Samsung chaebol includes semiconductors, consumer electronics, ship-building, financial services and a whole lot more, and is run by the family of Lee Byung-Chull, the founder of Samsung. As we'll see below, the chaebol business system is intricately linked to Korea's emergence as a dominant player in the memory industry.

Easy Access to Capital
South Korea, which used to be an agricultural nation, started an industrialization effort in the 1960s. The Government used a "guided capitalism" policy for this, wherein chaebol were selected to undertake projects, and were channeled funds from foreign loans. The government guaranteed repayment should a company be unable to repay its foreign creditors. Additional loans were made available from domestic banks. Not just that, semiconductor divisions of Korean chaebol were often provided capital by more profitable divisions of the conglomerate. This easy availability of capital was particularly important for success in the memory industry, which needed costly fabs and had long periods of losses.

To illustrate this point, let's look at Samsung. The company's average capital expenditure over revenues between 1987-92 was 39.8%, nearly double the industry average of 21%. In this period, the company also invested more than twice the amount of money in DRAM R&D compared to any of its competitors. The large investment helped Samsung become the world's biggest DRAM producer by revenue in 1992. Easy access to loans was the reason why Samsung had so much money to invest - its debt-to-equity ratio was 323%, which was more than twice that of similar American corporations.

Quick Transition to Next Wafer Size
Korean DRAM producers such as Samsung were some of the first to invest in larger diameter wafers. This strategy paid rich dividends, since equipment vendors provide discounts to the first few (risk-taking) customers. Let me give you some specific examples. In the early 1990s, the industry had an option to move from 6 inch to 8 inch wafers. The period coincided with a recession in the DRAM market, and Japanese chip makers were particularly hesitant to invest in 8 inch capacity. Samsung, which had abundant capital, had no such concerns and was the second company in the industry (after IBM) to move to 8 inch wafers. It got a 15% discount in 8 inch equipment prices, and benefited early from the 1.8x higher productivity of 8 inch fabrication. In the highly competitive DRAM industry, such a price advantage was invaluable. Similarly, in the move from 8 inch to 12 inch wafers in the early 2000s, Samsung led the way. As a result, it got a 24% equipment price discount and benefited early from the 2.3x higher productivity of 12 inch fabrication. This helped the company extend its market share lead in the early 2000s. No surprises why Samsung is investing aggressively in 450mm... its following a strategy that has worked in the past.

Capitalizing on Early Stages of the Product Life Cycle
Figure 2 shows a typical DRAM life cycle... it reveals how a 512Mb DRAM was sold at $11 for its first few months after introduction but reduced in value to $3 within three years. To take advantage of high prices just after introduction, ramping products to high yield quickly was a focus for Korean companies. And they executed superbly on this goal. 80% yield numbers were sometimes reached just after manufacturing started. This was because Korean companies had a system wherein yield could be estimated and improved even in the development phase... (as you'd know, in the rest of the world, yield is frequently optimized only in the manufacturing phase)

Picture
Figure 2: Typical DRAM Life Cycle

Low Cost Manufacturing Location
Korea is a lower cost manufacturing location than the US and Japan, and this factor played an important role in its emergence. Corporate income tax in Korea is in the 25% range, compared to 35-40% in the US and Japan. Furthermore, the Korean government gave 5-10 year tax holidays and low interest rate loans to chaebol. Labor costs were significantly lower in Korea as well. In the highly competitive memory industry, these advantages were priceless.

There are a few more reasons why Korean companies became dominant in the memory industry, but I'll save them for later. Before we end, let me talk a bit about the future. As you'd know, Korean companies were not around in the 1970s and 1980s when basic ideas on DRAM and flash memory were developed... due to this, they often had to license core memory patents from other companies and pay royalties. To ensure that they have a good patent position in future memory technologies such as RRAM, PCM, MRAM and monolithic 3D, Samsung and Hynix have invested large amounts of money in research. Several breakthroughs from these companies have been presented in top semiconductor conferences recently. In particular, during my visit to Korea, I was happy to see 3D applied to many markets - right from construction to NAND flash memory. Yup, you read the last sentence right - I did say construction. Korea has some of the highest density of skyscrapers in the world. See Figure 3. The Koreans certainly believe the third dimension is the way to go!

Picture
Figure 3: Korean companies are aggressively applying 3D both to construction and to semiconductors
- Post by Deepak Sekar

References
[1] Creating first mover advantages - the case of Samsung Electronics, by J-S. Shin and S-W. Jang
[2] A Study of the DRAM Industry, by J. Kang
7 Comments
Mahidhar Rayasam
11/20/2011 09:55:43 pm

Excellent post again, Deepak! On a similar note, I wonder what led to Thailand becoming an equivalent hub for HDDs. We just saw how such concentration of production in one country can spell disaster for the entire supply chain

Reply
Deepak Sekar
11/20/2011 10:24:02 pm

Hi Mahidhar,

Thanks for the feedback. Yes, it is a bit of a risk having much of the world's supply in one location. Luckily, Korea is not earthquake country...

Regarding Thailand, I've heard the lower labor costs and favorable government policies helped move HDD manufacturing there. Here's a nice article on the subject: https://editorialexpress.com/cgi-bin/conference/download.cgi?db_name=serc2009&paper_id=183

Thanks,
Deepak

Reply
Balaji
11/20/2011 10:57:44 pm

Excellent article Deepak and nice question Mahidar.

Reply
Abhijit Athavale link
11/24/2011 06:17:54 pm

Hi Deepak,

Nice blog. I actually took your blog and wrote about how some of the concepts are applicable in India here: http://punechips.com/korea-as-a-memory-hub-and-india-as-a/

Please take a look. Your feedback would be appreciated.

-Abhijit

Reply
Deepak Sekar
11/24/2011 11:38:17 pm

Hi Balaji, Thank you for your nice feedback.

Hi Abhijit,

Thank you for sharing the nice article. I spent a number of years in Pune and did part of my schooling there, so reading a blog called "Pune chips" was great!

Regarding manufacturing in India, I had a couple of other people contact me about that. I feel the main challenges there are the fact that the government needs to have stable policies and should be prepared to give big incentives to companies to set up shop there. Unfortunately, we have changes of state/central government every 5 years and the new government does not always uphold policies started by the previous one :-(

I feel India has thus far been most successful in activities that rely minimally on the government, such as software or chip design. But hey, a cash rich company like Reliance with good government contacts and manufacturing experience may be able to pull it off, especially if they invest in the right people.

Reply
Rob Leachman link
3/2/2012 02:15:08 am

I am disappointed that no mention is made about Samsung's cycle time management. Until 1995, Samsung had the worst manufacturing cycle time of any of the DRAM manufacturers. By then end of the 1998, they had the best. They still do. This transformation profoundly changed the DRAM industry. Samsung could flood the market first, capturing high prices for themselves and driving down DRAM prices for everyone else. That led to the collapse of the Japanese DRAM companies and the Korean government's bailout of effectively bankrupt Hynix/LG.

For a technical overview of the cycle time management methodology, see http://www.ieor.berkeley.edu/People/Faculty/leachman-pubs/SLIM.pdf

Reply
Deepak Sekar
3/2/2012 02:18:09 am

Hi Rob,

Thanks for the great comment. Cycle time was not mentioned largely due to my ignorance. I appreciate your pointing it out.

Deepak

Reply

Your comment will be posted after it is approved.


Leave a Reply.

    Search Blog


    Meet the Bloggers


    Follow us


    To get email updates subscribe here:


    Recommended Links

    3D IC Community
    3D IC LinkedIn Discussion Group

    Recommended Blogs

    • 3D InCites by Francoise von Trapp
    • EDA360 Insider by Steve Leibson
    • Insights From the Leading Edge by Phil Garrou
    • SemiWiki by Daniel Nenni, Paul Mc Lellan, et al.

    Archives

    May 2023
    March 2022
    December 2021
    August 2021
    August 2018
    July 2018
    May 2018
    October 2017
    September 2017
    December 2016
    September 2016
    August 2016
    November 2015
    October 2015
    September 2015
    July 2015
    June 2015
    May 2015
    April 2015
    March 2015
    February 2015
    October 2014
    September 2014
    August 2014
    July 2014
    June 2014
    May 2014
    April 2014
    March 2014
    February 2014
    January 2014
    December 2013
    November 2013
    October 2013
    September 2013
    August 2013
    July 2013
    March 2013
    February 2013
    January 2013
    December 2012
    November 2012
    October 2012
    August 2012
    June 2012
    May 2012
    April 2012
    March 2012
    February 2012
    January 2012
    December 2011
    November 2011
    October 2011
    September 2011
    August 2011
    July 2011
    June 2011
    May 2011
    April 2011
    March 2011

    Categories

    All
    3d Design And Cad
    3dic
    3d Ic
    3d Nand
    3d Stacking
    3d Technology
    Brian Cronquist
    Dean Stevens
    Deepak Sekar
    Dram
    Education
    Heat Removal And Power Delivery
    Industry News
    Israel Beinglass
    Iulia Morariu
    Iulia Tomut
    Monolithic3d
    Monolithic 3d
    MonolithIC 3D Inc.
    Monolithic 3d Inc.
    Monolithic 3d Technology
    Moore Law
    Outsourcing
    Paul Lim
    Repair
    Sandisk
    Semiconductor
    Semiconductor Business
    Tsv
    Zeev Wurman
    Zvi Or Bach
    Zvi Or-Bach

    RSS Feed

© Copyright MonolithIC 3D Inc. , the Next-Generation 3D-IC Company, 2012 - All Rights Reserved, Patents Pending