carlsverre

create database kv;
use kv;

/*

possibly add a namespace to all of this
lets start with BLOB for values

GLOBAL:
exists key

carlsverre

DROP database if exists analytics;
CREATE DATABASE analytics;
USE analytics;

CREATE TABLE entities (
    ts DATETIME(6) NOT NULL,
    id BIGINT NOT NULL,
    kind TEXT NOT NULL,

    SORT KEY (ts, id),

carlsverre

DROP database if exists analytics;
CREATE DATABASE analytics;
USE analytics;

create table visitors (
    visitor_id BIGINT NOT NULL,
    props JSON NOT NULL,

    PRIMARY KEY (visitor_id)
);

carlsverre

#!/usr/bin/env bash
set -euo pipefail

METADATA_BASE="http://metadata.google.internal/computeMetadata/v1/instance/attributes"

log() {
    local msg="${*}"
    logger -p syslog.info -t "startup-script" -- "${msg}"
}

carlsverre

Logistics

• https://www.singlestore.com/blog/scaling-worldwide-parcel-logistics-with-singlestore-and-vectorized/
• https://github.com/singlestore-labs/singlestore-logistics-sim
• https://github.com/singlestore-labs/singlestore-workshop-data-intensive-app
• https://www.youtube.com/watch?v=EDQjy_BMSyQ

Gaming

• https://github.com/singlestore-labs/esports-low-latency-analytics
• https://www.singlestore.com/customers/stream-hatchet/

carlsverre

sdb-admin list-nodes --role aggregator -q \
  | xargs -I NODEID sdb-admin -y update-config --memsql-id NODEID --key http_api --value on
sdb-admin list-nodes --role aggregator -q \
  | xargs -I NODEID sdb-admin -y update-config --memsql-id NODEID --key http_proxy_port --value <port>
sdb-admin list-nodes --role aggregator -q \
  | xargs -I NODEID sdb-admin -y query --memsql-id NODEID --sql "RESTART PROXY"

carlsverre

This is an example of how you can use RBAC and the CURRENT_SECURITY_GROUPS function to implement data hiding in SingleStore.

1. Get a SingleStore free license key from https://portal.singlestore.com
2. Set your SingleStore license key as an environment variable

export LICENSE_KEY="PUT YOUR LICENSE KEY HERE"

3. Run SingleStore cluster in a box

docker run -it \

carlsverre

create database if not exists t;
create database if not exists t2;

use t;
create table if not exists x (id int);

delimiter //

create or replace procedure foo (i int) as
begin

carlsverre

1. Download process.py
2. Run

~/csp $ python3 process.py
usage: process.py aws|azure|gcp

~/csp $ python3 process.py aws | head -n 10
3.5.140.0
3.5.140.1

carlsverre

#!/usr/bin/env bash
set -eux

DEVICES=$(ls /dev | egrep -o nvme.n1 | sort -u | grep -v $(df -h | egrep -o nvme.n1) | sed 's_\<_/dev/_g')

echo "Setting up the following devices:"
echo ${DEVICES}

read -p "Are you sure? " -n 1 -r
echo