Chapter 6: Order Types & Matching Engine

The Deutsche Boerse T7 trading platform supports a sophisticated order management framework with over 15 distinct order types, multiple matching algorithms, and comprehensive risk protections at the order level. Understanding these mechanisms is critical for HFT developers building trading strategies on Eurex derivatives and Xetra equities markets.

This chapter examines the order types available across Eurex and Xetra, the matching algorithms that execute these orders, and the various protections and controls that operate at the order level to maintain fair and orderly markets.

Introduction to T7 Order Management

The T7 system processes orders from submission through execution or cancellation using a multi-layered approach:

Order Type Validation: Ensures the order conforms to allowed types and parameters for the specific instrument
Risk Controls: Applies pre-trade risk checks including maximum order value, quantity limits, and price reasonability
Matching Algorithm Selection: Routes the order to the appropriate matching logic based on product configuration
Book Management: Maintains order priority according to algorithm-specific rules
Execution Reporting: Delivers fill confirmations with full execution details

The T7 platform handles orders for multiple asset classes across Eurex (futures and options on equity indices, interest rates, commodities, dividends, volatility, and FX) and Xetra (German and international equities, ETFs, ETCs, bonds). Each product category may utilize different order types, matching algorithms, and trading phases.

Basic Order Types

T7 supports three fundamental order types that form the foundation for all trading activity:

Order Type	Description	Key Behavior	Restrictions
Market	Execute immediately at best available price with no price limit	Not visible in order book, prioritized over limit orders	Not allowed for complex instruments (strategies)
Limit	Execute at specified price or better	Visible in order book until executed or cancelled	Available for all product types with all time-in-force options
Stop	Inactive order that becomes active when trigger price reached	Converts to active order (market or limit) when stop condition met	Reference price configurable (last traded, best bid/ask)

Market Orders

Market orders provide certainty of execution but uncertainty of price. Key characteristics include:

Immediate Execution: Matched against the best available opposite-side limit orders
No Book Visibility: Not displayed in order book depth since they execute immediately
Priority: Executed before limit orders at the same price level
Fill Behavior: May be partially filled if insufficient liquidity at best price, remainder executes at next price level
Pro-Rata Products: Automatically flagged as Immediate-or-Cancel (IOC) for products using pro-rata matching

Market orders are particularly useful for strategies prioritizing execution certainty over price control, such as unwinding positions quickly or capturing immediate market opportunities.

Limit Orders

Limit orders represent the majority of order book activity and provide price certainty with execution uncertainty:

Price Protection: Never executes worse than specified limit price
Book Visibility: Displayed in order book with full or partial quantity (iceberg orders)
Time-in-Force Flexibility: Supports all TIF options including persistent orders that survive system restarts
Modification Rules: Price changes create new order, quantity decreases maintain time priority
Universal Support: Available across all T7 products including complex instruments

Limit orders form the foundation of liquidity provision and enable market makers to post two-sided quotes efficiently.

Stop Orders

Stop orders enable conditional trading strategies by remaining inactive until market conditions trigger them:

Trigger Mechanism: Becomes active when reference price reaches or crosses stop price
Reference Price Options: Last traded price, best bid (for buy stops), best ask (for sell stops)
Conversion Behavior: Transforms into market or limit order (stop-limit) upon activation
Use Cases: Stop-loss protection, breakout strategies, trailing stops

The T7 system evaluates stop conditions continuously during trading phases and immediately converts triggered stops into active orders subject to normal matching rules.

Time-in-Force Options

Time-in-Force (TIF) instructions control order lifecycle and persistence:

TIF	Code	Behavior	Persistence	Key Use Cases
Good-for-Day	GFD	Expires at end of trading day	Non-persistent (deleted at market close)	Intraday strategies, daily reset positions
Good-Till-Cancelled	GTC	Active until executed or cancelled	Persistent (survives Market Reset, auto-deleted after 1 year)	Long-term limit orders, standing quotes
Good-Till-Date	GTD	Active until specified date	Persistent (survives Market Reset)	Event-driven strategies, calendar spreads
Immediate-or-Cancel	IOC	Execute immediately, cancel unfilled portion	Non-persistent	Liquidity taking, minimizing market impact
Fill-or-Kill	FOK	Fully executable or deleted entirely	Non-persistent (cash markets only)	All-or-nothing execution requirement

Persistence Behavior

Order persistence determines whether orders survive system maintenance events and session boundaries:

Persistent Orders (GTC, GTD):

Written to persistent storage (database)
Survive T7 Market Reset events
Restored automatically when trading resumes
Higher latency due to database write operations
Visible across all sessions
Automatically deleted after 1 year (GTC) or specified date (GTD)

Non-Persistent Orders (GFD, IOC, FOK):

Stored in memory only
Deleted during Market Reset
Lower submission latency
Must be re-entered after system restart
Visible across all sessions

Lean Orders:

Fastest submission path (memory-only, no replication)
Visible only to submitting session
Not survived across Market Reset
Optimal for HFT strategies with continuous order management
Available for both persistent and non-persistent TIF types

Type	Persistent	Visible to Other Sessions	Survives Market Reset	Latency	HFT Suitability
Standard Persistent	Yes (GTC/GTD)	Yes	Yes	Higher (DB write)	Medium
Standard Non-Persistent	No	Yes	No	Lower	High
Lean	No	No (submitting session only)	No	Lowest	Highest

Execution Restrictions

T7 provides several execution restrictions that modify how orders interact with the order book:

Book-or-Cancel (BOC)

Book-or-Cancel orders ensure passive liquidity provision by preventing immediate execution:

Behavior: Order rejected if immediately executable upon entry
Use Case: Market makers ensuring they only add liquidity, never take it
Availability: Options and futures products with closing auction support
Rejection: Generates specific error code when executable condition detected

BOC orders guarantee that submitted orders will rest in the book and not cross the spread, making them ideal for liquidity provision strategies that require rebate capture or regulatory liquidity requirements.

Market-to-Limit

Market-to-Limit orders provide hybrid behavior combining market order urgency with limit order price protection:

Initial Execution: Behaves like market order, executing against available liquidity
Conversion: After partial fill, converts remainder to limit order at last execution price
Price Protection: Prevents adverse selection on remainder after initial liquidity exhausted
Use Case: Large orders requiring immediate partial fill with price protection on remainder

This order type bridges the gap between full market orders (execution certainty but no price protection) and pure limit orders (price certainty but execution uncertainty).

Eurex Special Order Types

Eurex markets support specialized order types for derivatives trading:

One-Cancels-the-Other (OCO)

OCO orders combine limit and stop orders into a single logical order:

Structure: Limit order component + stop order component
Single Order ID: One Exchange Order ID for both components
Cancellation Logic: Execution of either component automatically cancels the other
Use Case: Bracket orders for profit-taking and stop-loss simultaneously
Priority: Both components maintain independent time priority in respective queues

OCO orders streamline order management by reducing message traffic and ensuring atomic cancellation when one side executes.

Trailing Stop

Trailing stops provide dynamic stop-loss protection that automatically adjusts to favorable price movements:

Variable Stop Limit: Stop price continuously adjusted based on reference price movement
Favorable Adjustment Only: Stop price moves only when market moves in profitable direction
Reference Price: Configurable (last traded, best bid, best ask)
Trailing Distance: Specified as absolute price offset or percentage
Activation: Becomes active order when final stop price reached

Trailing stops are particularly valuable for momentum strategies and protective stops that lock in profits while allowing positions to run.

Auction-Only Orders

Auction-only orders restrict execution to specific auction phases:

Opening-Auction-Only: Only executable during opening auction
Closing-Auction-Only: Only executable during closing auction
Rejection: Automatically cancelled if not executed during designated auction
Use Case: Index replication strategies, passive execution at auction prices

These order types enable strategies targeting auction prices specifically, such as closing price tracking or minimizing market impact through auction-only execution.

Strategy Orders (Multi-Leg)

Eurex supports complex strategy orders spanning multiple legs:

Strategy Type	Description	Max Legs	Use Case
Futures Combinations	Calendar spreads, butterfly spreads	Multiple	Term structure trading
Standard Options	Vertical spreads, straddles, strangles	Multiple	Volatility strategies
Non-Standard Options	Custom multi-leg combinations	Up to 5	Complex options strategies
Option Volatility	Volatility surface trading	Multiple	Vol arbitrage

Strategy orders execute atomically - all legs fill together or none fill, ensuring ratio integrity and preventing adverse selection on partial fills.

Xetra Special Order Types

Xetra cash markets introduced several specialized order types in December 2024:

Xetra Midpoint Orders

Midpoint orders enable hidden liquidity matching at the midpoint of the visible order book:

Order Type	Description	Execution	Transparency
Limit Midpoint	Limit order executed at midpoint if available	Midpoint only	No pre-trade transparency
Market Midpoint	Market order executed at midpoint if available	Midpoint only	No pre-trade transparency
Limit Sweep	Try midpoint first, forward to CLOB if no match	Midpoint then CLOB	CLOB portion visible
Market Sweep	Try midpoint first, forward to CLOB if no match	Midpoint then CLOB	CLOB portion visible

Midpoint Execution Details:

Execution price = (best_bid + best_ask) / 2
Requires tight spread (within configured maximum)
No pre-trade transparency (orders not visible in book)
Matching occurs before CLOB for sweep orders
Sub-tick price improvement for traders
Reduced market impact for large orders

Midpoint orders provide an alternative liquidity pool with price improvement for both buyers and sellers while maintaining market quality through spread requirements.

Iceberg Orders

Iceberg orders enable large order placement while limiting market impact:

Hidden Quantity: Only small portion (peak) visible in order book
Minimum Peak: At least 5% of total quantity must be visible (factor of 20 maximum)
Refresh Behavior: Hidden quantity automatically replenishes peak when executed
Time Priority: New peak receives new timestamp upon refresh (loses priority)
Use Case: Large institutional orders, minimizing information leakage

Example: 10,000-share iceberg with 500-share peak appears as 500 shares in book. After 500-share execution, next 500 shares appear with new timestamp.

Pegged Orders

Pegged orders automatically adjust their limit price relative to a reference price:

Reference Prices: Best bid, best ask, midpoint, last traded
Offset: Configurable price offset (positive or negative) from reference
Automatic Adjustment: Price updates continuously as reference moves
Priority: Maintains time priority unless price change required
Use Case: Algorithmic trading, relative value strategies

Pegged orders enable dynamic pricing strategies that track market movement without continuous manual repricing.

Minimum Acceptable Quantity (MAQ)

MAQ provides fill-size protection for orders:

Behavior: Order only executable if minimum quantity available
Partial Fill Control: Prevents numerous small fills
Combination: Works with other order types (limit, midpoint, etc.)
Use Case: Large orders avoiding death by a thousand cuts
Rejection: Order not matched if available quantity below MAQ

MAQ helps institutional traders control execution quality by ensuring meaningful fill sizes.

Matching Algorithms

T7 employs three primary matching algorithms, each optimized for different product characteristics:

Price-Time Priority (PTP)

Price-Time Priority is the primary matching algorithm for most T7 products:

Algorithm Steps:

Price Priority: Orders at best price (lowest for sell, highest for buy) matched first
Time Priority: At same price level, earliest order timestamp matched first (FIFO)
Pro-Rata Within Time: If multiple orders share identical timestamp, pro-rata allocation

Key Characteristics:

Incentivizes early order placement for priority
Rewards liquidity providers willing to improve price
Simple and transparent priority rules
Modification behavior affects priority

Modification Rules:

Modification Type	Time Priority Impact	Rationale
Quantity Increase	Lost (new timestamp)	Prevents gaming by adding quantity for priority
Quantity Decrease	Maintained (original timestamp)	Allows partial cancellation without penalty
Price Change	Lost (treated as new order)	Price change represents new trading decision
Other Attributes	Maintained (original timestamp)	Non-price/quantity changes don’t affect queue

Product Coverage:

All Xetra equities and ETFs
Most Eurex futures (equity index, interest rate, commodity)
Eurex options excluding equity options
Complex instruments (strategies)

Price-Time matching provides predictable execution priority and encourages tight spreads through price competition.

Pro-Rata (PTR) Matching

Pro-Rata matching allocates incoming order quantity proportionally among resting orders at the best price:

Algorithm Steps:

Price Priority: Best price matched first (same as PTP)
Pro-Rata Allocation: At same price, quantity allocated proportionally to displayed size
Size Priority: Larger orders receive proportionally more allocation
Time Priority: After pro-rata rounding, residual quantity allocated by time

Allocation Formula:

allocated_qty = min(order_open_qty, floor((order_qty / total_best_price_qty) * incoming_qty))

Example:

Order Book (Best Price Level):
Order A: 100 shares (earliest)
Order B: 300 shares
Order C: 600 shares
Total: 1,000 shares

Incoming: 500-share market order

Allocation:
Order A: floor(100/1000 * 500) = 50 shares
Order B: floor(300/1000 * 500) = 150 shares
Order C: floor(600/1000 * 500) = 300 shares
Total: 500 shares (no residual)

If incoming = 501 shares:
Residual 1 share allocated to Order A (time priority)

Key Characteristics:

Incentivizes larger order size for better fill rates
Reduces incentive to race for time priority
Encourages displayed liquidity over hidden orders
Market orders automatically flagged IOC for pro-rata products

Product Coverage:

All Eurex equity options
Selected Eurex products configured for pro-rata

Pro-rata matching is particularly suitable for markets where displayed liquidity is valued and racing for time priority is less desirable.

Pro-Rata with Top-of-Book Priority

This hybrid algorithm combines time priority for the top of book with pro-rata allocation for the rest:

Algorithm Steps:

Price Priority: Best price matched first
Top-of-Book Priority: First order at best price receives allocation until fully filled
Pro-Rata for Rest: Remaining orders at same price allocated pro-rata
Time Priority for Residual: After pro-rata, residual by time

Example:

Order Book:
Order A: 200 shares (earliest)
Order B: 300 shares
Order C: 500 shares
Total: 1,000 shares

Incoming: 500-share market order

Allocation:
Order A: 200 shares (filled completely - top of book priority)
Remaining: 300 shares for pro-rata

Order B: floor(300/800 * 300) = 112 shares
Order C: floor(500/800 * 300) = 187 shares
Residual: 300 - (112 + 187) = 1 share → Order B (time priority)

Final:
Order A: 200 shares
Order B: 113 shares
Order C: 187 shares

This hybrid approach balances time priority incentives for aggressive quoting with pro-rata benefits for displayed liquidity.

Auction Matching

Auctions employ specialized matching logic optimized for single-price execution:

Auction Types:

Opening Auction: Determines opening price and initial execution
Closing Auction: Determines closing price (widely used as reference price)
Intraday Auction: Scheduled or event-driven auctions during trading day
Volatility Auction: Triggered by price movements outside dynamic corridors

Price Determination Objectives:

Maximize Tradeable Volume: Select price executing most shares
Price Continuity: Prefer price closest to reference price (last traded, prior auction)
Market Pressure: If tie, prefer price favoring market buy/sell pressure

Auction Process:

Call Phase: Orders entered, modified, cancelled (order book accumulation)
Price Determination: Calculate auction price based on objectives
Order Book Freeze: Brief freeze period (product-specific)
Execution: All matched orders execute at single auction price
Publication: Auction result published, continuous trading resumes

Key Characteristics:

Single-price execution for all participants
Uncrossing ensures no two executable orders remain
Supports order entry, modification, cancellation during call phase
Indicative auction price may be published during call phase
Surplus orders remain in book after auction

Auctions concentrate liquidity at specific times and provide transparent price discovery, making them ideal for index replication, large orders, and benchmark pricing.

Self-Match Prevention (SMP)

Self-Match Prevention prevents a trading firm’s own orders from matching each other, avoiding wash trades and potential regulatory issues:

Purpose:

Prevent business unit orders from matching internally
Avoid wash trades (artificially inflating volume)
Mandatory for proprietary algorithmic orders under MiFID II (effective April 1, 2021)

Implementation:

SMP Type A on T7 platform
Identification via Tag 28744 (MatchInstCrossID) in ETI protocol
Active during continuous trading phase only (not auctions)
Configurable cancellation behavior

SMP Cancellation Options

As of Release 14.0 (December 1, 2025), T7 provides three cancellation options when self-match detected:

Option	Code	Behavior	Default Since	Use Case
Cancel Passive	CP	Resting order cancelled, incoming order matched	Dec 1, 2025	Prioritize new order intent
Cancel Aggressive & Passive	CAP	Both orders cancelled	Pre-Dec 2025	Maximum wash trade prevention
Cancel Aggressive	CA	Incoming order cancelled, resting order remains	Optional	Prioritize existing quotes

Default Change: Prior to December 1, 2025, the default was Cancel Aggressive & Passive (CAP). The new default of Cancel Passive (CP) reflects market participant feedback preferring to prioritize incoming order intent while maintaining wash trade prevention.

Configuration:

Set at Business Unit level
Applied consistently across all orders from that unit
Configurable via ETI session parameters

Example Scenario:

Situation:
- Order A: Sell 100 @ 50.00 (resting in book, Business Unit X)
- Order B: Buy 100 @ 50.00 (incoming, Business Unit X)
- Both orders have same MatchInstCrossID

Outcome by SMP Setting:
- CP (Cancel Passive): Order A cancelled, Order B matches with other liquidity
- CAP: Both Order A and Order B cancelled
- CA (Cancel Aggressive): Order B cancelled, Order A remains in book

Self-Match Prevention is essential for algorithmic trading firms running multiple strategies that may generate opposing orders inadvertently.

Passive Liquidity Protection (PLP)

Passive Liquidity Protection addresses latency arbitrage and stale quote sniping by deferring aggressive order execution:

Purpose:

Protect market makers from latency-based arbitrage
Prevent stale quote sniping when market makers updating quotes
Level playing field for liquidity providers
Encourage tighter spreads and deeper order books

Mechanism:

Aggressive orders (taking liquidity) delayed by product-specific deferral time before matching
Passive orders (posting liquidity) impact order book immediately without delay
Deferral applied at matching engine before order matching logic

PLP Deferral Times by Product

Product Category	Deferral Time	Implementation Date
EURO STOXX 50 Index Options (OESX)	1 ms	August 2019 (pilot)
DAX Index Options (ODAX)	1 ms	August 2020
STOXX 600 Index Options (OMSX)	1 ms	Subsequent rollout
Equity Options (single stocks)	1 ms	Ongoing expansion
Fixed Income Options	1 ms	Ongoing expansion
Money Market Index Options	1 ms	Ongoing expansion
FTSE 100 Options (OTUK)	3 ms	October 2022
FX Futures & Options	Varies by product	Product-specific

Order Type Behavior:

Order Type	Classification	PLP Applied
Market Order	Aggressive	Yes (deferred)
IOC Limit Order	Aggressive	Yes (deferred)
FOK Order	Aggressive	Yes (deferred)
Resting Limit Order (new)	Passive	No (immediate)
Resting Limit Order (existing)	Passive	No (immediate)

Matching Timeline Example (1 ms PLP):

T+0.00 ms: Aggressive market order arrives at matching engine
T+0.00 ms: PLP deferral timer starts
T+1.00 ms: Order released to matching algorithm
T+1.00 ms: Order matches against order book
T+1.05 ms: Fill confirmation sent to participant

During 0-1 ms window:
- Market makers can update/cancel stale quotes
- Other passive orders can be submitted
- Order book can adjust before aggressive order matches

Market Impact:

Tighter spreads observed post-implementation (DAX options: spread reduction 10-15%)
Increased order book depth (more displayed liquidity)
Reduced adverse selection for market makers
Minimal impact on overall market quality or execution speed

Rollout History:

August 2019: Pilot for EURO STOXX 50 options (OESX)
August 2020: DAX options (ODAX) after successful pilot
October 2022: FTSE 100 options (OTUK) with 3 ms deferral
Ongoing: Expansion to additional equity options and other asset classes

PLP represents a significant market structure innovation addressing the latency arms race while maintaining market quality and fairness.

Risk Controls at Order Level

T7 implements several order-level risk controls to prevent erroneous orders and protect market integrity:

Maximum Order Value (MOV)

Maximum Order Value provides fat finger protection by rejecting excessively large orders:

Implementation:

Calculated as: Transaction Size Limit × Contract Size × Tick Value
Applied at order entry before order reaches matching engine
Effective limit = min(user-configured limit, exchange-defined limit)
Rejection generates specific error code for immediate feedback

Limit Determination:

Exchange sets per-product maximum based on typical trading volumes
Annual review and adjustment by Deutsche Boerse risk committee
Participants can set lower limits via risk management parameters
Different limits for different product categories

Example Calculation:

Product: ODAX (DAX Options)
Contract Size: EUR 5
Tick Value: EUR 0.10
Transaction Size Limit: 10,000 contracts

Maximum Order Value: 10,000 × 5 × 0.10 = EUR 5,000
Maximum Notional: Depends on strike price and option value

MOV prevents single erroneous orders from causing extreme market dislocations while allowing legitimate large orders within reasonable boundaries.

Maximum Order Quantity

Product-specific quantity limits prevent unreasonably large orders:

Implementation:

Hard limit per product category
Rejection at order entry with specific error code
Published in product specifications
Can be more restrictive than MOV for certain products

Example Limits:

Product Type          | Max Quantity
----------------------|-------------
Equity Index Options  | 50,000 contracts
Fixed Income Futures  | 10,000 contracts
Single Stock Options  | 25,000 contracts
Commodity Futures     | 5,000 contracts

Quantity limits complement MOV by preventing extreme position concentrations regardless of notional value.

Price Reasonability Checks

Price collars prevent orders with clearly erroneous prices from entering the order book:

Implementation:

Entry Interval: Acceptable price range around reference price
Reference Price: Last traded, last auction, theoretical price (product-dependent)
Rejection: Orders outside interval rejected immediately
No Trading Halt: Unlike volatility interruptions, price collar violations do NOT trigger trading halts

Price Collar Characteristics:

Check Type	Trigger	Action	Market Impact
Price Collar	Order price outside entry interval	Reject order	None (no halt)
Volatility Interruption	Potential execution outside dynamic corridor	Halt trading, auction	Temporary halt

Application Scope:

EnLight (OTC clearing): Prevents unreasonable clearing prices
TES (Trade Entry Service): Validates manually entered trades
Regular Order Entry: Applied to standard limit orders

Price reasonability checks provide an early filter for erroneous prices without disrupting continuous trading.

Volatility Interruptions

Volatility Interruptions pause continuous trading and transition to auction mode when price movements exceed configured thresholds:

Dynamic Price Corridors

Dynamic corridors establish acceptable price ranges based on recent market activity:

Trigger Conditions:

Incoming order would execute outside dynamic corridor range
Reference price: last traded price or theoretical price
Corridor width: configurable percentage based on product volatility
Fast Market Adjustment: Corridors auto-widen during “fast market” declaration

Corridor Parameters:

Corridor Width = Reference Price × Volatility Factor
Upper Limit = Reference Price × (1 + Volatility Factor)
Lower Limit = Reference Price × (1 - Volatility Factor)

Example (5% corridor, Reference = 100.00):
Upper Limit = 100.00 × 1.05 = 105.00
Lower Limit = 100.00 × 0.95 = 95.00

Fast Market Behavior:

Exchange declares fast market based on heightened volatility
Corridor width automatically increases (e.g., 5% → 10%)
Prevents excessive interruptions during high-volatility periods
Automatic reversion to normal corridors when conditions normalize

Static Price Corridors (Release 12.1+)

Static corridors complement dynamic corridors by preventing full-day price dislocations:

Implementation:

Reference prices: Last auction price, last traded price, theoretical price
Corridor width: Wider than dynamic corridors (typically 10-20%)
Objective: Prevent circuit breaker triggers from extreme single-day moves
Complementary: Both dynamic and static corridors evaluated

Static vs. Dynamic Comparison:

Aspect	Dynamic Corridor	Static Corridor
Reference Price	Recent traded price (updates frequently)	Auction price, daily open (updates less)
Width	Narrow (2-5%)	Wide (10-20%)
Purpose	Prevent rapid short-term moves	Prevent extreme single-day moves
Update Frequency	Continuous	Per auction or session
Priority	Primary protection	Secondary protection

Automated Corridor Expansion (ACE) - Xetra

Xetra’s Automated Corridor Expansion widens corridors progressively when auctions fail to produce valid prices:

ACE Process:

Volatility Interruption triggered
Volatility Auction begins
If no valid price: Wait minimum interval (120 seconds single-level, 120-300 multi-level)
Automatically widen corridor to next level
Repeat until valid price found or Extended Volatility Interruption

Expansion Levels:

Level 1: Standard corridor (e.g., 5%)
Level 2: First expansion (e.g., 8%)
Level 3: Second expansion (e.g., 12%)
Level 4: Final expansion (e.g., 20%)

If no valid price at Level 4: Extended Volatility Interruption (manual intervention)

Interval Requirements:

Single-Level Products: Minimum 120 seconds per level
Multi-Level Products: 120-300 seconds depending on product configuration
Ensures sufficient time for participants to reassess and adjust orders

Volatility Interruption Process

Full VI workflow from trigger to resolution:

Trigger Event:

Scenario: Eurex ODAX product
Current best bid: 18,500
Current best ask: 18,510
Dynamic corridor: ±3% (555 points)

Incoming order: Buy at 19,100
Potential execution: 19,100 (outside upper limit 19,055)
Action: Volatility Interruption triggered

VI Workflow:

Immediate Halt: Continuous trading suspended instantly
Instrument Status: Changed to “Volatility Interruption” (broadcast via market data)
Transition to Auction: Volatility Auction call phase begins
Order Management: Participants can enter, modify, cancel orders during call phase
Indicative Price: T7 publishes indicative auction price (if calculable)
Price Determination: Auction matching algorithm calculates final price
Validation: Check if auction price within valid corridors
Resolution Scenarios:
- Valid Price: Auction executes, continuous trading resumes
- No Valid Price (Xetra): ACE widens corridor, repeat auction
- No Valid Price (Final): Extended Volatility Interruption (manual intervention required)
Resumption: If successful, return to continuous trading mode

Frequency Limits:

No minimum interval between consecutive VIs (removed in recent releases)
Allows rapid successive interruptions during extreme volatility
Each interruption treated independently

Market Data:

VI trigger broadcast immediately
Instrument status updates via EOBI and EMDI
Indicative auction price published during call phase
Final auction result published before continuous trading resumes

Volatility interruptions balance market safety (preventing disorderly trading) with efficiency (minimizing trading halt duration).

Order Book Management

Order Priority After Modification

Understanding priority rules is critical for optimizing order placement and modification strategies:

Modification Type	Time Priority	Price Level	Rationale
Quantity Increase	Lost (new timestamp)	Same	Prevents gaming via quantity addition for priority
Quantity Decrease	Maintained (original timestamp)	Same	Allows partial cancellation without penalty
Price Improvement	Lost (new timestamp)	New (better)	New price = new order entry
Price Worsening	Lost (new timestamp)	New (worse)	New price = new order entry
TIF Change	Maintained (original timestamp)	Same	Non-execution attribute change
Other Attributes	Maintained (original timestamp)	Same	Display quantity, identifiers, etc.

Strategic Implications:

Quantity reductions safe for maintaining priority (used by market makers adjusting quotes)
Any price change loses priority - better to cancel and replace if speed critical
Quantity increases lose priority - submit separate order instead for HFT strategies
Non-price/quantity modifications preserve queue position

Iceberg Order Refresh Behavior

Iceberg orders balance large order execution with market impact minimization:

Initial Entry:

Total Quantity: 10,000 shares
Peak Quantity: 500 shares (5% of total)
Hidden Quantity: 9,500 shares

Order Book Display:
Order ID 12345: 500 shares @ 50.00 (timestamp T1)

After Peak Execution (500 shares):

Execution: 500 shares @ 50.00
Remaining Total: 9,500 shares
New Peak: 500 shares (next tranche)
Hidden Quantity: 9,000 shares

Order Book Display:
Order ID 12345: 500 shares @ 50.00 (timestamp T2 - NEW)

Priority Lost: New timestamp T2 moves order to back of queue at 50.00

Strategic Considerations:

Peak size determines visibility vs. priority tradeoff
Smaller peaks = lower market impact but more frequent priority loss
Larger peaks = fewer refreshes but higher information leakage
Minimum 5% (20:1 ratio) enforced on Xetra

Market Reset Synchronization

Market Reset events require careful state synchronization:

Pre-Reset:

Participants receive Market Reset notification
Countdown period before reset execution
Opportunity to cancel non-persistent orders

During Reset:

T7 system brief downtime (typically <1 minute)
Persistent orders written to disk
Non-persistent orders deleted
Session state cleared

Post-Reset:

System restart and connection re-establishment
Persistent orders restated via Extended Order Information messages
Participants process restatement messages to synchronize state
Lean orders deleted (not restated)
Trading resumes after synchronization period

Synchronization Requirements:

Participant State Before Reset:
Order A: GTC, 1000 shares @ 50.00, Order ID 12345
Order B: GFD, 500 shares @ 51.00, Order ID 12346
Order C: Lean GTC, 200 shares @ 49.00, Order ID 12347

Post-Reset Expected State:
Order A: Restated, same parameters, possibly new Order ID
Order B: Deleted (GFD non-persistent)
Order C: Deleted (Lean not restated)

Required Action:
- Match restated Order A by original ClOrdID
- Confirm deletion of Order B and Order C
- Re-enter Order B and Order C if still desired

Proper Market Reset handling is essential for avoiding duplicate orders and maintaining accurate position tracking.

Summary

The T7 order management framework provides a sophisticated toolkit for implementing diverse trading strategies:

Order Type Selection:

Basic types (market, limit, stop) cover fundamental needs
TIF options control lifecycle and persistence
Execution restrictions (BOC, market-to-limit) fine-tune behavior
Specialized types (OCO, trailing stop, midpoint) enable advanced strategies

Matching Algorithm Understanding:

Price-Time for most products incentivizes early placement
Pro-Rata for equity options rewards size and displayed liquidity
Auctions concentrate liquidity at specific times for price discovery
Algorithm choice affects optimal order placement strategies

Risk Management:

Order-level controls (MOV, quantity limits, price collars) prevent errors
Self-Match Prevention avoids wash trades
Passive Liquidity Protection addresses latency arbitrage
Volatility interruptions maintain orderly markets

Priority Optimization:

Modification rules determine queue position preservation
Iceberg refresh behavior trades visibility for market impact
Pegged orders enable dynamic pricing without manual updates

For HFT developers, mastering these order management mechanisms is foundational to strategy implementation, risk control, and latency optimization on Deutsche Boerse markets.

Previous: Chapter 5 - Market Data Feeds

Table of Contents

Next: Chapter 7 - Market Models & Microstructure