Reduce Data Movement and Decrease Processing Times with a Machine Scale Feature Store(Molecula)


Manage episode 293297170 series 2881014
Travis Lawrence tarafından hazırlanmış olup, Player FM ve topluluğumuz tarafından keşfedilmiştir. Telif hakkı Player FM'e değil, yayıncıya ait olup; yayın direkt olarak onların sunucularından gelmektedir. Abone Ol'a basarak Player FM'den takip edebilir ya da URL'yi diğer podcast uygulamalarına kopyalarak devam edebilirsiniz.

In this episode, we speak with H.O. Maycotte. H.O. is the CEO/founder of Molecula, an enterprise feature store that simplifies, accelerates, and controls big data access to power machine-scale analytics and AI. Molecula is powered by Pilosa, an open source project created by H.O. and team. Pilosa eliminates the need to copy data between systems in order to make it accessible for analytical and machine learning purposes at scale. Leading companies like Spotify, Hulu, Uber, Zillow, ESPN are all utilizing Pilosa.

Top 3 Value Bombs:

  1. Large volume or complex workloads on a lake or DW can take hours to process, feature stores can reduce that to seconds in some scenarios.
  2. On average, over 80% of an organization's data are copies of the original data. Reduce the movement and duplication of data across your organization.
  3. Organizations do not have a data volume problem but rather data readiness problem. Enable the business with “real time” data.

24 bölüm