
Sign up to save your podcasts
Or


With all of the drama associated with AI, it's easy to miss the need to understand the foundations that deliver the data that is the raw element from which AI value is built. Databases and storage infrastructure are critical components that have to work in concert with AI plans and returning guests Henry Baltazar and James Curtis join host Eric Hanselman to discuss what's been happening and what enterprises need to know about the future. Databases and storage management systems have been intertwined for a long time and AI pressures are tightening that connection. Storage systems perform analytics on the data they store to optimize its handling, tracking use and characteristics. The same insights that aid in compression and tiering are also useful in classifying data for AI. Data classification has always been a challenge for enterprises, as storage systems are typically disconnected from the data owners and applications that use them. Intelligent storage systems have been able to intuit the nature of content, including mapping databases and virtual machines. Databases have been able to leverage storage capabilities like snapshotting for resilience.
Into this mix a new set of AI focused storage and database offerings arrive that target AI uses. The question is whether the native database and storage systems can do enough of what's needed. They already store key data and have valuable insights and classification capabilities. Some vendors are attaching GPU clusters to storage systems to provide high performance AI model training functionality. The major issue for most, is the matter of data placement. Shifting petabytes of data is no small task and concerns about data security and the costs involved now loom much larger.
More S&P Global Content:
For S&P Global Subscribers:
Credits:
By S&P Global Market Intelligence4.9
2828 ratings
With all of the drama associated with AI, it's easy to miss the need to understand the foundations that deliver the data that is the raw element from which AI value is built. Databases and storage infrastructure are critical components that have to work in concert with AI plans and returning guests Henry Baltazar and James Curtis join host Eric Hanselman to discuss what's been happening and what enterprises need to know about the future. Databases and storage management systems have been intertwined for a long time and AI pressures are tightening that connection. Storage systems perform analytics on the data they store to optimize its handling, tracking use and characteristics. The same insights that aid in compression and tiering are also useful in classifying data for AI. Data classification has always been a challenge for enterprises, as storage systems are typically disconnected from the data owners and applications that use them. Intelligent storage systems have been able to intuit the nature of content, including mapping databases and virtual machines. Databases have been able to leverage storage capabilities like snapshotting for resilience.
Into this mix a new set of AI focused storage and database offerings arrive that target AI uses. The question is whether the native database and storage systems can do enough of what's needed. They already store key data and have valuable insights and classification capabilities. Some vendors are attaching GPU clusters to storage systems to provide high performance AI model training functionality. The major issue for most, is the matter of data placement. Shifting petabytes of data is no small task and concerns about data security and the costs involved now loom much larger.
More S&P Global Content:
For S&P Global Subscribers:
Credits:

1,897 Listeners

2,675 Listeners

1,639 Listeners

1,095 Listeners

1,836 Listeners

6 Listeners

1,448 Listeners

40 Listeners

9 Listeners

6 Listeners

676 Listeners

228 Listeners

28 Listeners

29 Listeners

10 Listeners

4 Listeners

64 Listeners

30 Listeners

11 Listeners

9,907 Listeners

4 Listeners

5,472 Listeners

189 Listeners

1 Listeners

60 Listeners

169 Listeners

139 Listeners

5 Listeners

3 Listeners

0 Listeners

6 Listeners

5 Listeners

5 Listeners

24 Listeners