Sep 22, 2014 · 21m · mad
Vance Loiselle, Sumo Logic // Data Driven #29 // Sep 2014 (Hosted by FirstMark Capital)
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
Sumo Logic CEO Vance Loiselle delivers a presentation at Data Driven NYC on how cloud-native machine data analytics and machine learning transform vast, unstructured enterprise log data into real-time operational, security, and business insights.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Matt holds 1.6% of the talking time here. How this is scored →
Matt as informed peer 0.3 Guest teaching 0.3 Guest disagreement 0.7 Matt pushing back 0.3
speaking balance: gold is Matt, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 15:25 Vance attacks key competitor Splunk
Vance explicitly targets competitor Splunk, declaring that Sumo Logic is taking them down hard through cloud-native innovation.
Hardest push from Matt ▶ 16:06 Matt Turck playfully presses Vance on Splunk commentaryHost Matt Turck jokingly calls out Vance's aggressive competitive rhetoric by asking him how he really feels about Splunk.
Biggest teaching moment ▶ 16:32 Vance educates on cloud adoption in enterprise environmentsVance reframes the question about cloud readiness by pointing out that major institutions like Bloomberg and financial firms are already heavy cloud users.
Matt holds his own ▶ 16:18 Matt Turck prompts Vance on big data cloud architectureMatt Turck confirms technical architecture details about cloud processing and presses Vance on overall market readiness for cloud big data.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Matt as informed peer | Guest teaching | Guest disagreement | Matt pushing back | Why |
|---|---|---|---|---|---|---|
| The Explosion of Machine Data | 0 | 0 | 0 | 0 | In this solo presentation segment, Vance Loiselle outlines the massive explosion of unstructured machine data generated by enterprise applications, streaming platforms, and IoT hardware like aircraft engines. The host is not active during this monologue. | |
| What Machine Data Can Tell You & Business Service Analytics | 0 | 0 | 0 | 0 | Vance demonstrates Sumo Logic dashboard capabilities and explains how unstructured log data powers business service analytics, application performance, and security monitoring. This is a monologue presentation without host involvement. | |
| Key Characteristics of a Machine Data Solution | 0 | 0 | 0 | 0 | Vance details the key technical pillars of Sumo Logic's architecture, emphasizing cloud elasticity, multi-tenant cost savings, real-time indexing, and machine learning data science. The monologue format entails zero host score metrics. | |
| How Sumo Logic Works and Scales | 0 | 0 | 0 | 0 | Vance explains the data ingestion pipeline, including lightweight collectors, 10-12x compression, real-time indexing, and scalable querying capabilities. The host remains silent during the presentation. | |
| Anomaly Detection, LogReduce, and Competitive Landscape | 0 | 0 | 3 | 0 | Vance explains anomaly detection and signatures before aggressively taking aim at competitor Splunk, claiming Sumo Logic is taking them down hard. His combativeness is directed at market competition rather than the host or audience. | |
| Q&A Session with Matt Turck and Audience | 2 | 2 | 1 | 2 | Matt Turck kicks off Q&A by teasing Vance about his stance on Splunk and asking clarifying questions on cloud processing and market readiness before handing the mic to audience members. The tone throughout remains collegial and inquisitive. |
Statements from this episode (17)
Sumo Logic has 150 employees and backing from Greylock, Sequoia, Accel
“Just real background, you can see where some of us came from are about a 150 people. We're backed out of the gate by Greylock, and then we just added, ah, Sequoia recently, and Excel, and Sutter Hill, and some other companies as investors.”
Netflix generates 10 to 30 terabytes of log data daily
“Netflix is spitting out anywhere from 10 to 20 or 30 terabytes a day”
A twin-engine airplane generates 40 terabytes of log data hourly
“One airplane, two engines is generating 40 terabytes of log data an hour.”
Global unstructured data generation will hit 20 million petabytes annually by 2016
“Within the next two years, the amount of unstructured data that will be generated is somewhere in the twenty million petabytes a year figure.”
One terabyte of log data equals roughly two billion events
“One terabyte is roughly about two billion events. It's two billion unique records.”
Midsize enterprises generate over two billion log records daily
“So even a midsize enterprise every day is generating over two billion records.”
Log data holds most of its actionable value within 15 minutes
“The majority of this data, the value is within 15 minutes of it being generated.”
Sumo Logic scaled to 400 customers in four years via AWS
“Having an idea and going a gray lock to now having 400 customers in four years was not possible. You know, even 10 years ago without companies like AWS and software and so on.”
A third of Sumo Logic's engineering team are data scientists
“About a third of our engineering team are data science folks who basically build algorithms to figure out what is common about this data across hundreds of customers”
Sumo Logic manages roughly 60,000 data collectors
“We probably have about 60,000 collectors under management today.”
Sumo Logic ingests 15 terabytes of compressed data daily
“So we're about, we're at about 15 terabytes of compressed data per day that comes into the system.”
Sumo Logic processes over 800,000 daily queries across four petabytes
“We do over 800,000 queries a day in our platform. So any given day, we have basically over four petabytes of data that's scanned in the platform and over 15 trillion records that people are going through.”
Sumo Logic is 'taking down Splunk hard'
“So, you know, we have one competitor out there. I'll name them. They're called Splunk. So if you're familiar with Splunk, we're taking them down, and we're taking them down hard, and they know it.”
Majority of mission-critical enterprise data is already cloud-hosted
“The majority of mission critical data that's out there today is in the cloud and people don't realize it.”