Zero Learning Curve

Simple TOML Config

Clean. Declarative. Readable by anyone.
Configuration that fits in your head.

DeclarativeHot ReloadSecure Defaults

See the Difference

Traditional Approach (20+ lines)

vcl 4.1;

sub vcl_recv {
  if (req.http.Cookie) {
    set req.http.Cookie = regsuball(
      req.http.Cookie,
      "(^|;\s*)(__utm[a-z]+|_ga|_gid)=[^;]*",
      ""
    );
    if (req.http.Cookie ~ "^\s*$") {
      unset req.http.Cookie;
    }
  }
}

sub vcl_backend_response {
  if (beresp.status == 200) {
    set beresp.ttl = 1h;
    set beresp.grace = 24h;
  }
  if (beresp.http.Cache-Control ~ "no-cache") {
    set beresp.uncacheable = true;
  }
}

sub vcl_deliver {
  if (obj.hits > 0) {
    set resp.http.X-Cache = "HIT";
  } else {
    set resp.http.X-Cache = "MISS";
  }
}

• Custom scripting language

• Imperative logic

• Regex-based patterns

• Specialized knowledge required

Trident TOML (8 lines)

# trident.toml

[cache]
default_ttl = "1h"
grace_period = "24h"

[cache.key]
strip_cookies = ["__utm*", "_ga", "_gid"]

[response_headers]
x_cache = true

• Standard TOML format

• Declarative config

• Readable patterns

• Any developer can read it

Why TOML Wins

📖

Readable by Anyone

Junior devs, ops teams, even managers can review and understand the config.

Zero Training
🔒

Secure by Default

Built-in cookie bypass, auth handling, and request smuggling prevention.

No CVE Risk
🔄

Hot Reload

Change config, save file, done. No restarts, no SIGHUP signals needed.

Instant Updates
✅

Validation on Load

Config errors caught immediately with clear error messages. No runtime surprises.

Fail Fast
📝

Git-Friendly

Simple diffs, easy code review. Compare to 500-line VCL changes.

Clean PRs
🚀

Faster Onboarding

New team members productive in minutes, not weeks of VCL training.

10x Faster

Real-World Examples

Stale-While-Revalidate

3 lines vs 25 lines
# Trident
[cache]
grace_period = "24h"
# Varnish VCL
sub vcl_hit {
  if (obj.ttl >= 0s) {
    return (deliver);
  }
  if (std.healthy(req.backend_hint)) {
    if (obj.ttl + obj.grace > 0s) {
      return (deliver);
    }
  }
  return (miss);
}

Cache Tags

Built-in & Ready
# Trident - Built-in cache tags
# Tags extracted from X-Cache-Tags header
# Purge via REST API:
curl -X POST /admin/purge/tag \
  -d '{"tag": "product:123"}'
# Traditional approach
# Often requires enterprise tier
# Or custom implementation
# Manual ban list management
# Additional complexity

Backend Health Checks

5 lines vs 15 lines
# Trident
[[backends]]
name = "origin"
host = "backend.example.com"
health_check_path = "/health"
health_check_interval = "5s"
# Varnish VCL
probe health {
  .url = "/health";
  .interval = 5s;
  .timeout = 2s;
  .window = 5;
  .threshold = 3;
}
backend origin {
  .host = "backend.example.com";
  .probe = health;
}

Simpler AND Faster

You might think simple config means slower performance. Wrong. Trident's rules engine adds only 1.1% overhead vs 68.5% for equivalent VCL.

1.1%
Trident Rules Overhead
245K RPS with complex rules
68.5%
Varnish VCL Overhead
Same rules, more CPU

Complete Production Config

# trident.toml - Complete e-commerce config

[server]
listen = "0.0.0.0:80"
admin_listen = "127.0.0.1:8080"

[cache]
max_memory = "4GB"
default_ttl = "1h"
grace_period = "24h"

[cache.compression]
enabled = true
algorithm = "zstd"
level = 3

[cache.coalesce]
enabled = true
timeout = "30s"

[cache.stale_if_error]
enabled = true
max_stale_age = "1h"
status_codes = [500, 502, 503, 504]

[cache.key]
vary_headers = ["X-Customer-Group"]
strip_cookies = ["__utm*", "_ga", "_gid", "fbclid"]
normalize_query = true

[[backends]]
name = "origin"
host = "backend.example.com"
port = 443
tls = true
health_check_path = "/health"
health_check_interval = "10s"

[logging]
format = "json"
level = "info"

40 lines of readable TOML. Full production setup.

Equivalent Varnish VCL: 200+ lines

Start with Simple Configuration

Declarative TOML is included in all Trident plans.

Start Free TrialTalk to Sales